The report of the Committee on Serious Violent and Sexual Offenders, chaired by Lord MacLean, highlighted the need for an independent public body to ensure the effective assessment and minimisation of risk. The Committee reported to Scottish Ministers in 2000 and its work informed the Criminal Justice (Scotland) Act 2003. The legislation introduced a new sentence, the Order for Lifelong Restriction (OLR) and established the Risk Management Authority (RMA).
This is the fourth edition of RATED, a directory intended to assist practitioners in selecting and applying risk assessment tools. Whether used as aids to initial decision making or case management in wider criminal justice settings, risk assessment instruments must be applied with an understanding of their respective strengths and weaknesses. In addition, to be used with any degree of confidence, risk instruments should have a sound empirical basis and validation history. The RMA’s ‘Framework for Risk Assessment Management and Evaluation’ (FRAME) highlights the importance of how this information is analysed, evaluated and then communicated in a meaningful way to inform decision making and action.
The Equalities Act (2010) has informed the latest edition of RATED. The Equalities Act maintains that the equalities duty of the public sector is to recognise differences based on social characteristics and to thereafter tailor the needs of individuals accordingly, to ensure the outcomes are equal for everyone in society. As part of its equalities duty, the RMA (2015) will “inform practitioners of relevant equalities considerations and help them apply appropriate risk assessment tools to specific populations.” Where the empirical research is available, the applicability of validated risk assessments to the relevant protective characteristics are considered: people with a mental impairment/learning disability (mental disability); people from an ethnic minority (ethnic group); females (gender); adolescents (age).
RATED is an ongoing project to be continuously updated and revised. Feedback and suggestions about new and emerging tools or new research studies related to the tools in the directory are welcomed.
© Risk Management
RATED page updated: September 2019 Authority 2019
1.1 Background to RMA
1 Background
1.2 Risk assessment tools
As part of its duties, the RMA is responsible for preparing and issuing guidelines on the assessment and minimisation of risk; as well as publishing standards by which measures taken in respect of the assessment and minimisation of risk are judged. In 2006, the RMA published the ‘Standards and Guidelines for Risk Assessment,’ focusing primarily on practice in relation to the Order for Lifelong Restriction (OLR). A year later, the ‘Standards and Guidelines for Risk Management’ was published.
Risk assessment instruments are intended to assist the practitioner in different ways by anchoring the assessment in empirical evidence to identify relevant risk factors. They may be used for a variety of purposes: aiding decision making, planning treatment, predicting the risk of recidivism, using in specialist settings or for case management in criminal justice settings. It is important that risk assessment instruments are applied with an understanding of their respective strengths and weaknesses. Fazel and Wolf (2017) highlight that external validation, performance in the population to be assessed and sound methodology are pivotal factors in deciding whether a tool should be chosen. In addition, to be used with any degree of confidence, risk instruments should have a sound empirical basis and validation history.
The publication of RATED is intended to assist practitioners apply appropriate tools as part of a structured approach to risk assessment to assist with the identification of risk factors, needs and strengths of an individual. It should be noted that the application of a tool is one step in the risk assessment process; it is essential that the emerging information is analysed, evaluated and then communicated in a meaningful way to inform decision making and action.
The evolution of risk assessment has been well documented, with the general recognition that the incorporation of the ever growing research literature delivers an incremental improvement to each new generation of risk assessment instrument (Andrews, Bonta and Wormith, 2006). Actuarial instruments are described as nondiscretionary procedures, whereby the evaluator makes a decision based on fixed and explicit rules designed to predict the future (Hart and Logan, 2011). Actuarial instruments were designed specifically to assess risk or estimate the probability or likelihood of an outcome using ‘predictor variables’ (Hart, Douglas and Guy, 2016).
RATED page updated: September 2019 © Risk Management Authority 2019
It2016).must
Whilst they incorporate the empirical evidence, they also include consideration of other clinical factors and do not lead to a quantified ‘score.’ In this sense, instruments can assist with the identification of risk factors and strengths of an individual (Baird 2017). There are a range of SPJ approaches, ranging from those requiring scenario planning and formulation to others that require an appraisal of the risk and protective factors of individual cases. Several instruments are primarily, although not exclusively, developed for application with mentally disordered individuals who offend; these are most commonly used by mental health professionals, psychologists and psychiatrists. Dynamic instruments allow for reassessments to be carried out to evaluate the effectiveness of treatment in order to guide future decisions. Each SPJ tool focuses on a particular offence type, such as sexual or adolescent violence, and, as such, these instruments are predominantly utilised in specialist settings for the purposes of detailed and individualised risk management planning (Abbiati et al., 2014; Lloyd, 2019; Logan
A further suite of risk assessment instruments, commonly known as ‘structured professional judgement (SPJ) instruments, are distinct from actuarial tools in that they are guidelines designed to reflect the discipline with respect to scientific knowledge and practice (Hart and Logan, 2011).
RATED page updated: September 2019 © Risk Management Authority 2019
The development of actuarial instruments has evolved over thirty years. Initially, instruments were based on static risk factors that may contribute to broad classifications based on longer term risk. These have since progressed to more sophisticated approaches that contain a range of dynamic factors and are designed to assist in an understanding of an individual and his/her behaviour, with the aim of reducing the likelihood of that behaviour through appropriate interventions (Hanson, 2007).
be highlighted, however, that concerns have been raised about an overreliance on the findings of risk instruments in proceedings that have a bearing on sentencing and/or restriction of liberty (Singh, Grann and Fazel, 2011). The implications from this are that all types of risk instruments share a common weakness if relied upon solely to evaluate ‘individual’ risk.
RATED page updated: September 2019 © Risk Management Authority 2019
In line with the conclusions of researchers (Boer, 2006; Rettenberger et al., 2009), the RMA Standards and Guidelines encourage a ‘convergent’ approach to risk assessment which incorporates the use of both actuarial and SPJ tools. Furthermore, structured professional judgement and decision making are necessary components of all risk assessment practice, regardless of the type of instruments employed or the professional background of the practitioner. It is within this context that the RATED has been developed.
Running in parallel with this development has been an on going academic debate concerning the applicability of group data to the individual and the relative superiority of actuarial and clinical approaches (Quinsey et al., 2006). A useful guide for practitioners is an approach that bridges both perspectives, drawing upon research work highlighting the limitations of the actuarial approach (Cooke and Michie, 2010; Hart, Michie and Cooke, 2007); while endorsing the counsel of others that abandonment of the actuarial tradition is unwise (Craig and Beech, 2010).
The RMA is tasked with promoting effective risk assessment and management practice. In July 2011, the RMA, along with Scottish criminal justice agencies and the Scottish Government, published the FRAME policy paper. This established the policy direction in which a shared consistent framework to promote proportionate, purposeful and defensible risk practice could be achieved. FRAME proposes an approach that is grounded in the principles of risk practice to ensure all professionals are aware of the processes and skills associated with effective risk assessment and management. It also helps professionals understand the contribution that different kinds of instruments make to this practice. Moreover, it firmly promotes the practitioner as the ‘assessor’ who skilfully, knowledgably and within the parameters of his/her competencies, applies appropriate tools in a structured, meaningful process. Logan (2016) recommends that risk judgment can be substantially structured by utilising formulation, which organises the information into an explanation for why it happened and the circumstances in which it could possibly transpire again.
RATED AIMS AND OBJECTIVES
This is the fourth edition of RATED, a guide intended to assist practitioners in applying risk assessment tools as part of the FRAME approach to risk assessment to assist with the identification of risk factors and strengths of an individual. RATED also intends to keep researchers updated about the most recent developments.
2
2.1 FRAME Approach
2.2 Purpose of RATED
The framework promotes risk assessment practice that makes meaningful use of risk assessment tools without being over-reliant on them, by ensuring that the valuable contribution of such instruments is located within a structured approach to risk assessment which recognises the strengths and limitations of such tools and the importance of professional/clinical judgement. Risk assessment tools do not necessarily provide the comprehensive range of risk, responsivity and protective factors that inform an understanding of individual risk. Extrapolating from this, these tools should be used to inform rather than replace professional/clinical judgement.
RATED page updated: September 2019 © Risk Management Authority 2019
• General Risk Assessment: those tools that can be used for general application for risk.
•To provide relevant research information on each assessment tool included in the directory, which includes (but is not limited to) validation evidence.
• Responsivity Issues: this category consists of instruments for individuals with personality disorders or learning disabilities.
• Youth Violence: instruments for measuring risk of violence in adolescents.
•To provide an impartial and factual account of the strengths and limitations of each instrument.
• Intimate Partner Violence and Stalking: instruments that relate to intimate partner violence and/or stalking.
The publication of RATED aims to provide a summary of the empirical evidence to inform a balanced and individualised approach to assessment, as well as to contribute to effective and ethical practice (Andrews and Bonta, 2010). The RATED directory is intended as a resource for Accredited Risk Assessors undertaking assessments for the High Court in Scotland when the court is considering imposing an OLR, in addition to other practitioners involved in risk assessment and risk Themanagement.objectives
• Youth General Assessment: similar to the General Risk Assessment category, but for use with adolescents.
• Violence Risk: instruments to establish the risk of violence.
RATED page updated: September 2019 © Risk Management Authority 2019
•To provide a guide for assessors to consider when applying a tool as part of a holistic risk assessment process.
• Youth Sexual Violence: tools designed for measuring sexual violence risk in adolescents.
• Sexual Offending: tools relating to sexual violence risk or internet offending.
Risk assessment instruments may focus on specific types of risk developed and/or particular groups. This directory is structured by type of tools. This introductory section provides further background to the purpose of RATED and the criterion used to rate tools. The categories of tools are the following:
of the current publication are as follows:
Whilst every effort has been made to ensure the accuracy of the information presented in this publication, given the evolving nature of research relating to risk assessment and risk management, the evidence base is subject to change.
Practitioners are, therefore, encouraged to keep abreast of the emerging evidence when administering the assessment tools. Additionally, practitioners are invited to contribute to the development of the evidence base and subsequent editions of this publication by providing feedback on the contribution of specific instruments to their practice by emailing RATED@rma.gov.scot.
The RMA’ s (2018a) revised Standards and Guidelines highlights the importance to practitioners of ensuring that risk assessment and management practices do not disadvantage those with ‘protected characteristics.’ This is entrenched within the Equalities Act (2010), which attributes ‘equalities duty’ upon public sector organisations. The purpose of this is to recognise differences based on a number of protected characteristics and tailor services accordingly to ensure that people are not disadvantaged on the basis of those characteristics. The protected characteristics are:
Ethnicity Age Mental disability Religion Gender Gender reassignment Sexual orientation
Hubbard and Pealer (2009) found that the more ‘issues’ an individual has, the less likely a programme will be successful in achieving what it aims to do. To that end, strategies that are informed by responsivity issues are more likely to be effective and reduce recidivism (Beaver and Schwartz, 2016).
RATED page updated: September 2019 © Risk Management Authority 2019
The ‘risk need responsivity principles,’ suggest that effective interventions are proportionate to the likelihood of re offending, address the needs related to offending, and are responsive to a range of issues that may influence a person’s ability to respond to interventions (Andrews and Bonta, 2010; Taxman et al., 2006).
2.3 Responsivity Considerations
Physical disabilities
RATED page updated: September 2019 © Risk Management Authority 2019
Pregnancy
1 The other protected characteristics within the UK Equalities Act (2010) are: religion, sexual orientation, marriage/civil partnership, pregnancy, gender reassignment (both biological changes or social functioning ones) and physical disabilities.
The research evidence in relation to the remainder of protected characteristics1 is lacking, with only a handful of studies available. For instance, transgender issues are explored in relation to the youth sexual violence tool MEGA♪ (Miccio Fonseca, 2018); whilst a study by Bhutta and Wormith (2016) incorporated a measure of religiosity and spirituality into the LS/CMI to test its effectiveness in a religious country.
When compiling a new edition of RATED, the available published literature relating to instruments was reviewed, including studies that have examined their applicability to specific groups. Where the empirical information is available, the applicability of risk assessments to the relevant protected characteristics is considered. Of the ten protected characteristics, a number have limited if any research evidence. Those that have attracted most attention in the literature are: people with a mental impairment/learning disability (mental disability) people from an ethnic minority (ethnic group) females (gender) adolescents (age)
Marriage/civil partnership
As part of its equalities duty, the RMA (2015) seeks to “inform practitioners of relevant equalities considerations and help them apply appropriate risk assessment tools to specific populations.” Since RATED is a means by which the RMA advances this objective, the equality and diversity guidance laid out in the Standards and Guidelines (RMA, 2018a) informs the content of RATED. An Equality Impact Assessment (EQIA) was also carried out to consider the role of protected characteristics in the publication of RATED.
In the application of tools, practitioners should be aware of the available research, or lack thereof, in relation to applicability to persons with a protected characteristic.
2.4 Developments in RATED Fourth Edition
•‘Internet Offending’ (offences related to indecent images of children) has been grouped with ‘Sexual Violence Risk’ and the category has been renamed as ‘Sexual Offending.’
Rearrangement of the directory entries:
•Learning Disabilities and Diagnostic/Personality Assessments were previously in separate categories. These four categories have now been amalgamated into a new category entitled ‘Responsivity Issues,’ documenting those tools specific to disorders that could influence offending.
Regrouping of reference lists:
RATED page updated: September 2019 © Risk Management Authority 2019
Reference lists are included as a separate document for each of the eight sections of tools. To facilitate accessibility for users, references within the lists are grouped by tools.
The existing research base is limited to the extent that many tools do not have published studies pertaining to protected characteristics. Moreover, when protected characteristics are reviewed in relation to risk assessment tools, the focus tends to be broad, e.g. mental ability broadly is reviewed; yet specific mental disorders receive less attention. If further research emerges then future editions of RATED may create new categories for these protected characteristics.
A number of significant changes have been made to the fourth version of the directory. These include a range of amendments to the format, layout and content of the directory, a summary of which is provided below.
•The Stalking category has been combined with the ‘Domestic Violence’ category, since these types of offences can often be interlinked (see Kerr, 2018). Additionally, the term ‘Domestic Violence’ has been replaced with the more commonly used term within the field ‘Intimate Partner Violence’ to give the new category of ‘Intimate Partner Violence and Stalking.’
Assessment, Intervention and Moving On Project Version 3 (AIM3), The AIM Project Assessment Models for Children under the age of 12 years old (AIM Under 12s), Protective and Risk Observations for Eliminating Sexual Offence Recidivism (PROFESOR) and Technology Assisted Harmful Sexual Behaviour: Practice (TA HSB Guidance), all in Youth Sexual Violence Risk.
Brief Spousal Assault Form for the Evaluation of RISK (B-SAFER) in the Intimate Partner Violence and Stalking section.
Oxford Risk of Recidivism (OxRec) tool in the General Risk Section.
RATED page updated: September 2019 © Risk Management Authority 2019
A number of tools have been upgraded from awaiting validation status in the third edition of RATED (RMA, 2015) to validated in this edition:
Terrorist Radicalization Assessment Protocol 18 (TRAP 18) and Workplace Assessment of Violence Risk (WAVR-21) in the Violence Risk section.
AssetPlus in Youth General Risk Assessment Section.
Comprehensive Assessment of Psychopathic Personality (CAPP) in the Responsivity Issues section.
Risk for Sexual Violence Protocol (RSVP) in the Sexual Offending section.
Upon reviewing the literature for the current risk assessment tools, a number of new tools were identified. Following a review of their usefulness and applicability, the decision was made to include the majority of them in this revised edition. The additional entries within the directory reflect both the extent of the review for this edition and the speed of development in the field of risk assessment. The new additions to this edition of RATED are:
Oxford Mental Illness and Violence Tool (OxMIV) in the Responsivity Issues section.
Structured Assessment of Positive Factors for Violence Risk: Youth Version (SAPROF:YV) and Short Term Assessment of Risk and Treatability: Adolescent Version (START:AV) in the Youth Vioelnce Assessment section
Inclusion of new tools:
Barr 2002R, Child Pornography Offender Risk Tool (CPORT), Sexual Offender Treatment Intervention and Progress Scale (SOTIPS) and Vermont Assessment of Sex Offender Risk 2 (VASOR 2) in the Sexual Offending section.
RATED page updated: September 2019 © Risk Management Authority 2019
In cases where the full article is already freely available on the internet (for instance, through ResearchGate) the hyperlink to this is included to facilitate users’ access to research. In a small number of cases, the research cited is not available online and, hence, no hyperlink is included.
The rating criteria refers to the amount of evidence available for a particular element (e.g. inter-rater reliability). The terminology has been changed in this version of RATED:
The Multiplex Guided Inventory of Ecological Aggregates for Assessing Sexually Abusive Adolescent and Children (MEGA) in the Youth Sexual Violence section.
•For zero bars, the wording of ‘no or poor’ has been changed to ‘insufficient.’
•For three bars, the word ‘little’ has been replaced with ‘preliminary.’
•For six bars, the term ‘some’ has been substituted with ‘intermediate.’
Hyperlinks to research:
The Youth Assessment and Screening Instrument (YASI) in the Youth General Risk section.
•For nine bars, the word ‘high’ has been replaced with ‘sufficient.’
To improve the accessibility to research, where available hyperlinks have been included to the studies cited in tool entries. The hyperlink will generally take the user to the journal’s page, which contains the article’s abstract. Users may then check whether they have access to the full article through their institution or they may have the option to purchase it on the journal’s landing page.
Hyperlinks are available within individual tool entries and in the list of references for each of the eight sections.
Rewording of rating criteria:
•Describing the purpose and design;
Where they have the necessary competencies and training, practitioners can use with confidence validated tools that possess a robust validation history and empirical grounding. These tend to be tools that have also evidenced sufficient inter rater reliability2, specificity3, sensitivity4 and predictive accuracy5 in identifying individuals at risk of re offending. Practitioners should use tools with caution that possess many of the essential attributes and have the potential to demonstrate all of these in the future (with further studies and/or evidence). Reports detailing findings based on the use of tools that are awaiting validation should outline the limitations of their use in this respect.
3 Specificity relates to the ability of a test to classify when an individual does not possess a particular characteristic.
•The age, gender, ethnicity, mental state and cognitive abilities of the individual;
•Evaluating the validation literature available;
4 Sensitivity refers to the extent to which an instrument can correctly classify when an individual possesses a particular characteristic. Sensitivity is inversely proportional with specificity, whereby one increases as the other decreases.
The use of suitable risk assessment tools is an important concern for practitioners and agencies involved in the management of offenders. RATED is designed to inform and support decisions regarding which tools to use by:
When using RATED, practitioners should consider the following:
•Providing information on the evidence base for the tool design;
Using RATED to select a risk assessment tool
2.5 Using RATED
RATED page updated: September 2019 © Risk Management Authority 2019
•The type of risk that they wish to evaluate (general risk of reoffending, risk of violence, sexual offences, intimate partner violence or stalking);
•Setting out the appraisal against the assessment framework.
•The performance of tools with respect to the criteria outlined which includes validation history, empirical grounding, inter rater reliability and ability to identify targets for intervention.
2 Inter rater reliability refers to the degree in which two or more assessors consistently rate items within a tool.
5 The predictive accuracy of a tool refers to its ability to distinguish between certain populations, such as those who reoffend and those who do not.
TheOLR.Standards and Guidelines states that practitioners should select risk assessment instruments appropriate to the risk needs and characteristics of the individual being assessed, as well as in accordance with their own competencies, experience and training. Practitioners should also be aware of the validity and
Practitioners should use tools with extreme caution that possess few of the essential attributes, but appear to have the potential to meet all of these in the future (with further studies and/or evidence). Reports detailing the findings based on the use of such tools must include a justification for selecting them and should outline the limitations of their use.
In cases involving the 17 21 year old age group, practitioners should use their judgement as to whether to apply instruments designed for young people, or those designed for an adult population. To inform this judgement, specific consideration should be given to the extent to which the tool incorporates evidence of its validity with young people and the impact of child developmental processes of the young person being assessed. Similarly, one should give careful consideration as to whether the selection of a given instrument is appropriate for certain minority groups, e.g. those with mental health problems, cognitive impairments and learning disabilities. A good guide to an instrument’s suitability for a particular group is whether a sample from the target group has been included in the original research sample or been the subject of a successful pilot study.
RATED page updated: September 2019 © Risk Management Authority 2019
In all cases, practitioners should have the necessary training for using the tool(s) of their choice; be aware of their limitations and the caveats of their use; be in a position to discuss these and to evidence their assessment. Importantly, these limitations should be clearly outlined in the assessor’s report to inform future decision making. Above all, risk assessment tools should be used as part of a risk assessment process to inform rather than replace professional judgement.
Practitioners are advised to review the range of guidance documents dealing with risk assessment. These include the RMA Standards and Guidelines, FRAME and the RMA accredited manner for risk assessment for Risk Assessment Orders (RAO) or Interim Compulsion Orders (ICO) when the High Court is considering an
limitations of any tool that they use (RMA, 2018a). RATED sets out the RMA’ s evaluation of the evidence available for each instrument and highlights any limitations or caveats that professionals should take into account when using a tool as part of a holistic risk assessment process.
The directory also highlights that there may be cases (perhaps with unusual types of offending) where available instruments have little or no validity, wherein the following activities are recommended:
•Use the structured professional judgement framework, as set out in this manner, to guide the process of case formulation and reaching a risk judgement;
RATED page updated: September 2019 © Risk Management Authority 2019
•Seek supervision from a sufficiently experienced and expert colleague;
•Seek the views of other experts on the approach to be taken in the case;
•Review the relevant publications to look at what the latest academic and clinical literature describes in relation to the type of offender and the approach to the risk under consideration;
•Gather as much information as possible to support the formal risk assessment;
•Be explicit in the report about the limitations of the approach used.
The RMA has developed a process for evaluating1 risk assessment tools and this has been followed for each tool presented in this document. This evaluation process will also be applied in future to assess newly developed tools. The RMA researcher working on this update to RATED used a standard pro-forma to describe each tool based on the summarised evidence and the adopted evaluation framework. This provides individual assessments of the tool against each of the framework criteria and comments to support them.
•Empirical Grounding
•Inter Rater Reliability
Compilation of evidence and evaluation
Evidence collated for the current audit, comprised of publications (including ‘grey literature,’ such as doctoral theses and dissertations for Master’s Degree and other unpublished material) and evidence of ongoing research.
•Validation History
1 In this document, the word ‘evaluation’ is used to refer to the assessment of the tools’ performance with respect to the criteria framework outlined.
The following criteria were used to select research studies for inclusion in the directory:
The RMA welcomes applications for inclusion in the directory, as well as any comments its users.
The adopted framework comprises a number of criteria considered essential for the evaluation of risk assessment tools, building upon earlier work by McIvor, Kemshall and Levy (2002) and the findings of other researchers (Heilbrun, Rogers and Otto, 2002; Rogers 2000). Criteria considered as essential are:
Adopting an evaluation framework
RATED page updated: September 2019 © Risk Management Authority 2019
3 EVALUATION PROCESS
3.1 The evaluation process
•An outline of its validation history, considering studies carried out in the United Kingdom and internationally; this includes details of the study focus and setting (community, secure unit, prison or mental health institution), the size of the sample and type of population used;
•Date of publication, with greater weight given to more recent publications;
•Focus of study, taking into account whether specific issues have been considered, such as predictive validity, inter rater reliability, significance/ usefulness,2 sensitivity and so on.
The following information was provided for each tool listed within the directory:
•Target population, i.e. male/female offenders, forensic mental clients, and children and young offenders;
•Size of study population, with greater emphasis on evidence drawn from large sample populations;
•The type of risk addressed, i.e. the risk of general, sexual and/or violent offending, or any combination of these;
•Place of publication, with greater weight given to publications in peer-reviewed journals;
•Authorship, i.e. whether the author(s) were, or were not, involved in the validation of the tool under consideration, with greater weight given to studies conducted by independent researchers rather than the author(s) of the risk assessment tool;
•A brief description of the risk assessment tool;
•Strengths, practice considerations, and future research directions.
2 Significance/Usefulness of a tool refers to its evidential basis, predictive power and clinical utility in enabling the practitioner to undertake risk formulation and risk management planning.
•Relevance to UK population; i.e. whether the study had employed a sample of individuals from the nations within the UK;
•Evidence of inter rater reliability, significance/usefulness;
RATED page updated: September 2019 © Risk Management Authority 2019
•Sponsorship, i.e. whether the study has been commissioned and/or funded by a government or statutory agency, with greater weight given to independent studies;
•The title and any acronym by which it is known;
•The publisher(s)/developer(s) and the year of publication (where known);
3.2 Criteria definitions and performance measurement scales
The RMA liaised directly with tool authors and publishers wherever possible to compile this material accurately.
The following criteria have been used to identify the strengths and weaknesses of each risk assessment tool listed within the current directory. It is important to note that these criteria have not been applied to risk assessment tools awaiting further validation. This is due to the limited research available relating to the psychometric properties of these risk assessment instruments.
RATED page updated: September 2019 © Risk Management Authority 2019
Empirical grounding examines the scientific and theoretical underpinnings of the risk assessment tools presented in this document. For example, a risk assessment tool based on sound theoretical evidence and / or other extensive scientific findings observed in prior research would be considered to have a high level of empirical grounding. Higher levels of empirical grounding may, therefore, increase the utility of the instrument in assessing the risk posed by an individual.
3.2.2 Inter-Rater Reliability
Tools that are awaiting validation currently do not have the sufficient research evidence to be considered validated; although these have the potential to become validated in future.
Tools are divided up into ‘Validated’ and ‘Awaiting Validation.’ Validated tools are those which have sufficient research studies evidencing their suitability to be used within the field. Due to the scope of research available on the, validated tools have more extensive entries than those that are ‘awaiting validation.’ A further distinction between these categories is that to rate a tool in as validated, the RMA considers the existence and quality of the validation evaluation studies, assessed on the basis of the availability of two or more independent papers written by different authors and published in peer reviewed journals.
Inter-rater reliability refers to the degree to which two or more assessors are consistent in their ratings when using the same risk assessment tool. There are two types of statistics used to estimate this: the Intra Class Correlation (ICC) and Kappa
3.2.1 Empirical Grounding
a.General Predictive Validity This item refers to the capability of the risk assessment tool to discern between the recidivist and non recidivist populations. The recognised statistical measure for this is the receiver operating characteristic curve (probability that a recidivist would receive a higher rating than a non recidivist) (Taxman, 2018).
(K)Coefficients (see section 5 for further details). Researchers must be able to demonstrate that the risk assessment instruments are reliable; without reliability results, the risk assessment’s performance cannot be replicated and validation cannot be attained. Even if a tool had predictive validity, its usefulness will be undermined if it is applied inconsistently by different practitioners (Taxman, 2018). It is thus of utmost importance for a risk assessment tool to have high inter rater reliability whereby assessors using the same tools score the items similarly.
To rate a tool in this area, the RMA considers the existence and quality of the validation evaluation studies, assessed on the basis of the availability of two or more independent papers written by different authors and published in peer reviewed journals. The papers are required to have examined the predictive validity of the tool and/or its practical usefulness for the assessment and management of risk of harm to others. This approach accommodates concerns that have been raised in the literature that different research designs may be appropriate for identifying the properties, strengths and limitations of various types of instruments.
The validation history criterion is split into four subsections which are as follows:
b.Applicability: Female Offenders This item refers to the validity of the risk assessment tool for women and girls.
d.Applicability: Mental Disabilities. This item refers to the validity of the risk assessment tool for those with mental disorders and/or learning disabilities.
3.2.3 Validation History
c.Applicability: Ethnic Minorities This item refers to the validity of the risk assessment tool for ethnic minority populations.
RATED page updated: September 2019 © Risk Management Authority 2019
Tools are rated on these criteria using the following performance scale:
evidence of sufficient empirical grounding, inter rater reliability or validation.
RATED page updated: September 2019 © Risk Management Authority 2019
3.2.4. Contribution to Risk Practice
0 Bars No evidence of or evidence indicating insufficient empirical grounding, inter-rater reliability or validation.
evidence of empirical grounding, inter rater reliability or 9validation.BarsSufficient
This section in tool entries supports the tiered approach to risk practice promoted within the RMA’s Framework for Risk Assessment, Management and Evaluation (FRAME), in particular the first three standards relating to: (1) Risk Assessment; (2) Planning and Responding to Change; (3) Risk Management Measures.
For tools that are ‘awaiting validation,’ this section is entitled ‘Tool Development’ and covers findings of studies in relation to predictive accuracy. ‘Tool Development’ will also cover the empirical grounding of a tool and studies relating to inter rater reliability where these are available.
3 Bars Preliminary evidence of empirical grounding, inter rater reliability or 6validation.BarsIntermediate
RATED page updated: September 2019 © Risk Management Authority 2019
3.2.5 Other Considerations
This final section documents other elements to be considered before using the tool, e.g. whether it has a clinical over ride feature to allow the risk rating to be readjusted at the assessor’s discretion. Also noted in this section are any other research findings about the tool that do not relate to its inter-rater reliability or predictive accuracy. In tools that are ‘awaiting validation,’ this section is entitled ‘General Notes’ and covers similar areas.
This section does not incorporate the ratings scales observed in the previous criterion; rather, it provides a qualitative evaluation of the properties of the instrument and its contribution to risk practice.
The Brier score ranges from 0 to 1. The best possible score is 0, indicating total accuracy. The lowest possible score is 1, meaning that the forecast was inadequate.wholly
Statistical Analyses
Definition Interpretation
The Brier score is defined as the average quadratic difference between the predicted probability and the binary outcome. Its purpose is to measure the ‘calibration’ of a set of predictions.probabilistic
There is a useful explanation that uses an analogy of the bull’s eye on a dartboard: actually hitting the bull’s-eye represents accuracy; landing shots together indicates good reliability. Considering this, hitting the bull’s eye and landing all the shots together would convey both accuracy and precision (Viera and Garrett, 2005).
4
Intra Class Correlation Coefficient poor/moderate/excellentdescriptors;(ICC)
An ICC correlationestimationrepresentsscoretheofthebetween two scores. It measures the magnitude of agreement for inter rater reliability.
RATED page updated: September 2019 © Risk Management Authority 2019
Authors of reports and studies on risk assessment tools adopt a range of descriptions of statistical outcomes. For the sake of consistency in the tables below, the RMA has adopted the following descriptors drawing from the most prevalently used terminology in the literature.
Brier Scores
STATISTICAL TERMINOLOGY
ICC values range from 0 to 1 and are typically reported with two decimal points, e.g. .75. Cicchetti followingrecommends(1994)thethresholds: <.40 = ‘poor’ .40 to .75 = ‘moderate’
–
The effect size quantifies the size of the difference between two groups. It is calculated as the standardised mean difference between the two groups: mean of Group A minus mean of Group B; the total of which is divided
Effect sizes
Effect sizes can be interpreted in terms of the percentiles or ranks at which two Interpretationsoverlap.distributions of effect sizes are dependent on the assumptions that the two groups are
Kappa (.) poor/average/excellentdescriptors;Coefficient
A Kappa is always less than or equal to a value of ‘1.’ A value of ‘1’ implies perfect agreement and a value of ‘ 1’ implies thresholdsRecommendeddisagreement.perfectare: 0 Less than agreementchance 0.01 0.20 poor agreement 0.21 agreementaverage0.60 0.61 agreementexcellent0.99
<
The Kappa Coefficient (.) measures the agreement between two individuals.
RATED page updated: September 2019 © Risk Management Authority 2019 .75 to 1.0 = ‘excellent’
normally distributed and have the same standard deviations.
The higher the value of sensitivity, the greater the ability of the measure being
Pearson small/moderate/largeCoefficientCorrelationdescriptors;
by the deviation.standard
Sensitivity is the ability of a test to correctly classify an individual as possessing a particular
RATED page updated: September 2019 © Risk Management Authority 2019
Pearson coefficientcorrelationmeasures the
The values of r can range from ‘-1’ to ‘1,’ with ‘0’ indicating that there is no .40.25.10follows:be(1988)Accordingrecidivism.decreasedassociatedscoresindicatenegativereincreasedassociatedhighindicatePositiveandpredictorbetweenrelationshipthevariabletheoutcome.valuesthatthescoresarewithcidivism;whereasvaluesthathigharewithtoCohen,rvaluesmayinterpretedasaresmall,aremoderate,arelarge.
association between a predictor variable and the outcome.
Sensitivity and Specificity
tested to correctly identify individuals.
In the context of risk assessment, the ROC curve is a plot that shows the probability
According to Cohen (1988), d values may be interpreted as .20follows:isconsidered ‘small’; .50 is considered ‘moderate’; .80 is considered ‘large’.
characteristic (e.g. offending). Specificity is the ability of a test to classify an individual as not possessing a particular characteristic (e.g. not decreases.increases,asspecificity,proportionalSensitivityoffending).isinverselywithmeaningthatthesensitivitythespecificity
Several ROCcalculatedindicatorsdifferentcanbefromcurves.The
Receiver CharacteristicOperating(ROC) Curve low/moderate/highdescriptors;
Z+ small/moderate/largedescriptors; Measures associationthebetween the predictor variable and the outcome. These two groups usually comprise of the (1) recidivists and (2)the non recidivists, separated by the difference in their scores obtained on assessments.risk
RATED page updated: September 2019 © Risk Management Authority 2019
For instance, a sensitivity of 62% on a risk assessment tool indicates that it has the ability to correctly classify just under two thirds of individuals who will reoffend. For specificity, higher values indicate the ability of a measure to correctly identify who will not possess characteristicscertain (e.g. who will not go on to reoffend).
RATED page updated: September 2019 © Risk Management Authority 2019
that a measure will correctly identify persons as recidivists or non recidivists (Mossman, 1994; Rice and Harris, 1995). It is a plot of the ‘hits’ (the proportion of recidivists correctly identified as recidivists) against the ‘false alarms’ (the proportion of identifiednon-recidivistsasrecidivists).
low/moderate/highdescriptors;
The area under the curve (AUC) for the ROC curve is a useful summary statistic for the extent to which a measure discriminates between recidivists and non recidivists.
most commonly used indicator is the ‘area under the curve’ (AUC) (see below).
AUC values can range from ‘0’ to ‘1’. They can interpretedbeas the probability that a randomly selected recidivist has a worse score than a randomly selected Valuesnon-recidivist.between ‘.51’ and ‘1.0’ indicate recidivism;associationspositivewithwhilst
Area Under the Curve (AUC)
values between ‘0’ and ‘.50’ indicate that predictions are no better than chance.
RATED page updated: September 2019 © Risk Management Authority 2019
Using Cohen’s (1988) d values as a guide, AUC values may be interpreted as .556follows:isconsidered ‘low’; .639 is considered ‘moderate’; .714 is considered ‘high.’
Andrews, D., Bonta, J. and Wormith, S. (2006) ‘The Recent Past and Near Future of Risk and/or Need Assessment.’ Crime & Delinquency 52(1), 7 27. Access Here
Craig, L. A. and Beech, A. R. (2010) ‘Towards a best practice in conducting actuarial risk assessments with adult sexual offenders.’ Aggression and Violet Behavior 15(4), 278 293. Access Here.
RATED page updated: September 2019 © Risk Management Authority 2019
Fazel S. and Wolf, A. (2018) ‘Selecting a risk assessment tool to use in practice: a 10 point guide.’ Evidence Based Mental Health 21(2),41 43. Access Here.
Cooke, D. J. and Michie, C. (2010) ‘Limitations of diagnostic precision and predictive utility in the individual case: A challenge for forensic practice.’ Law and Human Behavior 34(4), 259 274. Access Here.
Andrews, D. A. and Bonta, J. (2010) The Psychology of Criminal Conduct (5th edition). Cincinnati, Ohio: Anderson Publishing. Access Here.
Baird, C. (2017). A question of evidence part two. Summary and conclusion. Madison, WI: National Council on Crime and Delinquency. Access Here.
Bhutta, M. H. and Wormith, J. S. (2016) ‘An examination of a risk/needs assessment instrument and its relation to religiosity and recidivism among probationers in a Muslim culture.’ Criminal Justice and Behavior 43(2), 204 229. Access Here.
Abbiati, M., Laix, J., Gasser, J. and Malin, V. (2014) ‘Predicting physically violent misconduct in prison: a comparison of four risk assessment instruments.’ Behavioral Science and the Law 37(1), 61 77. Access Here
Boer, D. P. (2006) ‘Sexual offender risk assessment strategies: is there a convergence of opinion yet?’ Sexual Offender Treatment 1(2), 1 4. Access Here
Cohen, J (1988) Statistical Power Analysis for the Behavioral Sciences. New York: Routledge. Access Here.
Reference List
Hart, S. D., Douglas, K. S. and Guy, L. S. (2016) ‘The Structured Professional Judgment Approach to Violence Risk Assessment: Origins, Nature and Advances.’ In L. Craig and M. Rettenberger, M. (eds.) The Wiley Handbook on the Theories, Assessment and Treatment of Sexual Offending. Hoboken, New Jersey: John Wiley and Sons, 643 666. Access Here
RATED page updated: September 2019 © Risk Management Authority 2019
Heilbrun, K., Rogers, R., Otto, R.K. (2002).’ Forensic assessment: Current status and future directions.’ In Ogloff, J.R.P. (Ed.). Taking psychology and law into the twenty first century. New York: Kluwer/Plenum, 119 146. Access Here.
Hart, S., Michie, C. and Cooke, D. (2007) ‘Precision of actuarial risk assessment instruments: Evaluating the ‘margins of error’ of group v. individual predictions of violence.’ British Journal of Psychiatry 190(S49), S60 S65. Access Here.
Hanson, R. K. (2007, March) ‘How should risk assessments for sexual offenders be conducted?’ Paper presented at the Fourth Annual Forensic Psychiatry Conference, Victoria, British Columbia, Canada. [Not accessible]
Hart, S. D. and Logan, C. (2011) ‘Formulation of violent risk using evidence based assessment: the structured professional judgment approach.’ In D. Sturmey and M. McMurran (eds.) Forensic case formulation. Chichester, UK: Wiley Blackwell, 83 106. Access Here.
Lloyd, M. (2019) Extremism Risk Assessment: A Directory. Lancaster, UK: Centre for Research and Evidence on Security Threats. Access Here
Hubbard, D. J. and J. Pealer. (2009) ‘The Importance of Reponsivity Factors in Predicting Reductions in Antisocial attitudes and cognitive distortion among adult male offenders.’ The Prison Journal 89 (1), 79 98. Access Here
Logan, C. (2016) ‘Structured professional judgment: applications to sexual offender risk assessment and management.’ In Phenix, A. and Hoberman, H. (eds.) Sexual Offenders: Diagnosis, Risk Assessment and Management. New York: Springer, 571 588. Access Here
Risk Management Authority. (2018b) Literature Review a review of the risk posed by internet offenders. Paisley: RMA. Access Here.
Rice. M. and Harris, A. (1995) ‘Psychopathy, schizophrenia, alcohol abuse and violent recidivism. ‘International Journal of Law and Psychiatry 18(3), 333 342. Access Here.
McIvor, G., Kemshall, H. and Levy, G. (2002) Serious Violent and Sexual Offenders: The Use of Risk Assessment Tools in Scotland. Edinburgh: Scottish Executive Social Research. Access Here.
V. L., Harris, G. T., Rice, M. E. and Cormier, C. A. (2006) Violent Offenders: Appraising and Managing Risk (2nd Edition). Washington, DC: American Psychological Association. Access Here.
Risk Management Authority. (2018a) Standards & Guidelines Risk Assessment Report Writing. Paisley: RMA. Access Here.
Risk Management Authority. (2016) Standards & Guidelines for Risk Management. Paisley: RMA. Access Here
RATED page updated: September 2019 © Risk Management Authority 2019
Risk Management Authority. (2015) Equality Duty Progress Report. Paisley: RMA. Access Here
Rettenberger, M., Matthes, A., Boer, D. P. and Eher, R. (2009) ‘Prospective Actuarial Risk Assessment: A Comparison of Five Risk Assessment Instruments in Different Sexual Offender Subtypes.’ International Journal of Offender Therapy and Comparative Criminology 54(2), 169 186. Access Here
Mossman, D. (1994) ‘Assessing predictions of violence :being accurate about accuracy.’ Journal of Consulting and Clinical Psychology 62(4), 783 792. Access Quinsey,Here.
Miccio Fonseca, L. C. (2018, Spring). Sexually abusive youth who are transgender. Perspectives: California Coalition on Sexual Offending (CCOSO) Quarterly Newsletter, 1, 14 17. Access Here.
RATED page updated: September 2019 © Risk Management Authority 2019
Singh, J. P., Grann, M. and Fazel, S. (2011) ‘A comparative study of violence risk assessment tools: a systematic review and meta regression analysis of 68 studies involving 25, 980 participants.’ Clinical Psychology Review 31(3), 499 513. Access Here.
Taxman, F. Thanner, S. M. and Weisburd, D. (2006) ‘Risk, Need and Responsivity (RNR): It All Depends.’ Crime & Delinquency 52(1), 28 51. Access Here.
Taxman, F. S (2018) ‘Risk Assessment: where do we go from here?’ In J. P. Singh, D. G.Kroner, J. S. Wormith, S. L. Desmarais and Z. Hamilton (eds.) Handbook of Recidivism Risk/Needs Assessment Tools. John Wiley & Sons, Ltd: New Jersey, 271 284. Access Here.
Viera, A. J. and Garrett, J. M. (2005) ‘Understanding Interobserver Agreement: The Kappa Statistic.’ Family Medicine 37(5), 360 363. Access Here.
Rogers, R. (2000) ‘The uncritical acceptance of risk assessment in forensic practice.’ Law and Human Behavior 24(5), 595 605. Access Here.
Description
•There is a feature in the instrument allowing the assessor to override the initial risk level (Guay and Parent, 2017).
Assessors must possess advanced training, certification and experience in psychological assessment or a related discipline, or satisfactorily complete a training course certified by the publishers. Can be used by a large range of professionals including social work and probation services.
•It is designed to assist professionals in management and treatment planning in justice, forensic, correctional, prevention, and related agencies.
•Combines risk assessment and case management in a single assessment tool.
•The LS/CMI is normed on Canadian and North American probation and institutional populations for males and females. Supplementary norms provided for use in Singapore and the UK.
•Risk is categorised into five levels; ‘very low’, ‘low’, ‘moderate’, ‘high’ and ‘very high. ’
Age Appropriateness
Year 2004
Strengths
RATED page updated: July 2019 © Risk Management Authority 2019
Author / Publisher Andrews, Bonta and Wormith
•The LS/CMI is a measure of risk and need factors with a case management component.
Category General Risk Assessment (Validated)
16+ Assessor Qualifications
•It expands the traditional risk/need assessment instrument to a more comprehensive assessment by including non criminogenic needs, prison experience and responsivity considerations.
•Assessors are able to identify strengths in the individual and his/her circumstances.
•It includes five assessment sections, three summary sections and three case management sections. Section 1 (General Risk/Needs Section) consists of 43 items that are grouped into 8 subsections (the Central 8) Other assessment sections include: Specific Risk/Need Factors, Prison Experience: Institutional Factors, Other Client Issues: Social, Health and Mental Health, and Special Responsivity Considerations. Structural issues like poverty, race, gender and age are addressed in the ‘Responsivity Considerations’ section (Wormith and Bonta, 2018).
Name of Tool Level of Service/Case Management Inventory (LS/CMI)
•It allows for a professional override of risk level based on an assessment of strengths and specific risk factors that are not captured in Section 1.
RATED page updated: July 2019 © Risk Management Authority 2019
Inter Rater Reliability
•Rettinger and Andrews (2010) the LS/CMI attained moderate to high inter rater reliability estimates ranging from .65 for ‘financial problems’ to .91 for the composite general risk/need score in a sample of females.
None at present.
•The LS/CMI utilised normative data from circa 20, 000 females in inmate and community settings in four countries in order to address gender specific risk and responsivity issues., including mothering concerns, adult victimisation issues and protective strengths.
Empirical Grounding
•Labrecque et al. (2017) found that 21 items demonstrated good or strong levels of inter rater agreement with an ICC exceeding .60. Seven items, however, gave an inadequate level of consistency, with an ICC of below .50. Three of these items are in the ‘Companions’ section; the remaining four items are across the Antisocial pattern, procriminal attitudes/orientation, leisure and family/martial sections. It is suggested that training around these items should be strengthened to address this.
General Predictive Accuracy
None at present.
•Developed in part from the LSI R, a well validated tool with the developments informed by further research and consultation with practitioners. The LS instruments are grounded in the ‘General Personality and Cognitive Social Learning’ theory, maintaining that behaviour such as criminal conduct is learned from interactions with others. It is a general theory, advancing the argument that the Central Eight risk/need features apply across gender, age and race. (Andrews, Bonta and Wormith, 2004; Wormith and Bonta, 2018).
a)UK Research
Validation History
•Section 1 was informed by a re analysis of LSI R item data following research in Ontario, Canada that became called the LSI Ontario Version When the LSI R was developed, GPCSL theory had not fully matured. For instance, the Emotional/Personal subcomponents of the LSI R was overly concerned with general feelings of emotional distress and psychotic illness and underemphasised antisocial personality features. Grounding in the ‘Central Eight’ and the inclusion of a case management system formed the basis for the development of the LS/CMI (Wormith and Bonta, 2018).
a)UK Research
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
On average, logistic regression models predict that each 1 point increase in total risk score increases the odds of arrest by about 3 4% and the odds of incarceration by about 5 LS/CMI9%.risk scores are the strongest predictor of recidivism for both Day Report Centre clients and Department Of Corrections inmates. Using only LS/CMI
•In order to test the tool in Pakistan, Bhutta and Wormith (2016) translated the LS/CMI into the national language, Urdu, , and adjusted some of the items (e.g. education subscale modified to align with the lower literacy rate than that of Western countries). The revised tool was administered to 55 adults released on probation and found to be a useful indicator of recidivism. The only notable difference was that female probationers tended to score higher on the general risk/needs score, which the authors suggested could be a reflection of a cultural or gender bias.
•The LS/CMI was adapted into a French version by Guay (2016). This version was utilised in a study by Guay and Parent (2017) to assess 3646 individuals with a sentence of less than two years. Good predictive accuracy was shown for new arrests and new convictions (AUC range of .70 .77 and .72 .77 respectively), with the exception of new convictions for other crimes (AUC equal to .66).
•Wormith, Hogg and Guzzo (2012) moderate to large correlations observed between the General Risk/Need Score and general recidivism (.47), violent recidivism (.28) and sexual recidivism (.17).
•Gordon, Kelty and Julian (2015) tested the LS/CMI on individuals who offended in Australia and found that its total scoring yielded a significant although weak predictive utility with an AUC of .62. On this basis, the authors caution that the LS/CMI may not be the most suitable tool for measuring risk in Australia.
•Wormith et al. (2007) in a 10 year follow up sample of 61 adult males, LS/CMI attained moderate accuracy (AUC) for the prediction of recidivism (any new conviction (.65), non violent conviction (.62), violent conviction (.68) and any re incarceration (.69). This tool was unable to predict sexual convictions (.49).
b)International Research
•A study by Spence and Haas (2015) in West Virginia found risk scores are strongly predictive of recidivism, even when controlling for other factors such as age, gender, and ethnicity.
risk scores, it is possible to correctly predict recidivism in 60 70% of cases. The inclusion of other variables (i.e., age, race, etc.) increases predictive accuracy of recidivism
•Dyck and colleagues (2018) looked at the predictive validity of the LS/CMI with 136 Atlantic Canadian individuals. Whilst it was found to be a strong predictor of recidivism for males (AUC=.75), it was even better for females (AUC=.94) over an average of 3.42 year follow up period.
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
Applicability: Females
•Caldwell et al. (2018) evaluated predictive validity of 19, 344 probationer records in Nebraska over a five and a half year period. It was found that the LS/CMI predicted outcomes better for minorities (those who did not identify as White Europeans or were White but of Hispanic descent) than non minorities (White European Americans or those of non Hispanic descent). An experiment was conducted to explore whether officers showed prejudice in their scoring; findings showed total risk scores remained stable across different racial groups.
None at present.
•Andrews et al. (2012) the LS/CMI composite score had a mean AUC of .83 for recidivism across five different samples of females. With gender found to have a significant effect on the validity of substance abuse, this was controlled for in analyses, resulting in an AUC of .79.
•Rettinger and Andrews (2010) in a 57 month follow up, the LS/CMI was able to discriminate between different risk categories for females within prison and community settings. Of the 411 women in the study, the higher risk females were responsible for 74% of all new offences. The
a)UK Research
•Olver and Kingston (2019) found that the LS/CMI (called the LSI Ontario Version in this paper) predicted violent and general recidivism in the overall sample and among specific diagnostic groups (schizophrenia, anxiety disorders and mood disorders). Predictive accuracy for violence specifically was smaller although still significant, suggesting the need to use a violence specific tool in conjunction with this one.
RATED page updated: July 2019 © Risk Management Authority 2019
a)UK Research
Validation History
•Within a larger study, Girard and Wormith (2004) included a sub sample of 169 prison inmates with mental health problems (depression, psychosis, previous suicide threats/attempts). This sub sample was found to score significantly higher on the General Risk/Need total score than those without mental health issues (mean effect sizes of 21.95 and 19.48 respectively).
Applicability: Mental Disorders
b)International Research
LS/CMI general risk/need component generated AUC scores for both general and violent recidivism of .87 and .86 respectively.
•In a meta analysis involving samples of Aboriginal individuals, the central eight risk factors attained small to moderate mean effect sizes in relation to general recidivism ranging from .19 (family/marital) to .56 (criminal history). Smaller mean effect sizes were observed for violent recidivism (Gutierrez et al., 2013)
None at present.
None at present.
Validation History
b)International Research
Applicability: Ethnic Minorities
a)UK Research
•Andrews (1995) found that those diagnosed with mental disorders categorised as ‘high / very high risk’ on the LS/CMI had recidivism rates of up to 73% for any re offending compared to 17% who were rated as ‘very low’ risk.
•Wormith, Hogg and Guzzo (2015) applied the LS/CMI to 9692 Aboriginal and 24, 758 non Aboriginal individuals Predictive accuracy was demonstrated for both sets, with AUC scores in the range of .64 for Aboriginals and .74 for the remainder. The authors advise that assessors should consider special circumstances when carrying out interviews with Aboriginal individuals, such as cultural heritage, jargon and dialect, communication styles and relational expectations.
•When testing the internal consistency of the LS/CMI, Gordon, Kelty and Julian (2015) looked at how well items in the LS/CMI correlated with the overall score. Twelve items were removed as a result of this test. Further, it was highlighted that five of the items within the LS/CMI could be considered ‘double barrelled’ questions: for instance, asking about both youth and adult criminal history in the same item. It is, thus, suggested that these items are separated to allow them to be adequately measured in this sample (Australian individuals aged 18 67 completing community based sentences)
•Following on from this and similar findings from other studies, Wormith and Bonta (2018) highlight cautious use of the professional override. It is recommended that any ‘excessive’ use defined as more than 5% of cases should be quickly addressed to avoid prediction of recidivism being compromised.
Other Considerations
•Wormith, Hogg and Guzzo (2012) found that when assessors applied the override this tended to be to increase rather than decrease risk level. It was found that this reduced the predictive validity of the scale by excessively increasing risk. For instance. 263 individuals who committed sexual offences were initially in the low risk category and were thereafter overridden to medium, high or very high risk. Despite this, they actually recidivated (24.2% for medium risk; 19% for high risk; 5.9% for very high risk) at a lower rate than individuals deemed to be at low risk (31.7%).
•Assessors are given the chance to elaborate on factors which have been highlighted as a strength in the ‘General Risk/Needs’ section (Andrews, et al. 2004: 153).
•The LS/CMI aids the assessor in identifying risk, need and responsivity factors relevant to the individual’s likelihood of re offending and of other issues relevant to a holistic case management plan.
•The developers of the tool indicated that the LS/CMI provides a ‘gender informed’ assessment for risk, needs and responsivity issues; thus, it can be used across various settings without the need for separate gender specific assessments.
•LS/CMI has an ability to highlight the strengths of the individual. These are factors that would actively enable the individual to desist from further offending and enables assessors to provide further information on these strengths in relation to potential clinical override of the level of risk generated from Section 1.
•The clinical/professional override feature was the focus of Guay and Parent’s (2017) study of 3646 individuals who offended in Quebec. In 144 of 3646 cases, the clinical override was used to reduce the level of recidivism; whilst in 93 instances, the measure was used to increase it. It is concluded that the ‘upward overrides’ (i.e. increased level of threat) had greater predictive accuracy than ‘downwards overrides’ to a lower risk level. On this basis, it is recommended that further research is carried out on protective factors and the situations allowing for a ‘downward override.’ The results also showed that the clinical override feature decreased the predictive accuracy of the LS/CMI, apart from in cases of convictions for new crimes.
Contribution to Risk Practice
RATED page updated: July 2019 © Risk Management Authority 2019
•Literature also describes how the LS/CMI may be used in recommendations for sentencing (see Wolbransky et al., 2012). A study looking at the correspondence between presentence risk evaluations and sentencing outcomes of 165 individuals using the LS/CM, LSI_R and HCR 20 found that sentencing outcomes were associated with risk assessment scores. (Jung et al., 2015).
•Many of the factors identified within the assessment can act as targets for treatment/change and the tool can aid assessors in determining the level of monitoring and supervision required with regards to the formulation of case management plans
•The LS/CMI does not measure religiosity (religious belief or feeling) or spirituality. A study by Bhutta and Wormith (2016) in the highly devout country of Pakistan added a measure of religiosity to the LS/CMI to test if the inclusion of this improved the predictive accuracy of the instrument. They. however, concluded that the addition of this to the instrument is unlikely to improve its ability to predict recidivism.
•The LS/CMI was pilot tested as the LSI OR (Andrews, Bonta and Wormith, 1995) for a number of years prior to its publication in 2004.
RATED page updated: July 2019 © Risk Management Authority 2019
•Recent independent research (not author affiliated) has been conducted on LS/CMI. These include a study investigating the predictive validity of gang and non gang members (Guay, 2012); an examination of the need principle and effect of treatment on change of LS/CMI scores (Holliday, et al. 2012); and a comparison on LS/CMI of psychopathic and neuropathic (characterised by frontal lobe deficits and psychosis) individuals who committed homicide (Gilligan and Lennings, 2011).
16+ Assessor Qualifications
•Provides structured professional decision making in a way that is comprehensive and consistent regardless of the case presented (Campbell et al., 2009).
RATED page updated: July 2019 © Risk Management Authority 2019
Description
•Ability to discriminate risk across various outcome measures such as spousal abuse recidivism (Hendricks et al., 2006).
Category General Risk Assessment (Validated)
•LSI R is a 54 item actuarial tool of the individual’s attributes and their circumstances. It is designed to assess criminogenic risk and identify the needs of those who have offended (Watkins, 2011).
Author / Publisher Andrews and Bonta
•Information is collected via a semi structured interview, a review of case records and collateral verification (Wilson et al., 2016).
Strengths
•Normed on North American prison, parole and probation populations.
Name of Tool Level of Service Inventory Revised (LSI R)
•In addition to recidivism, composite scores help to predict parole outcomes and the presence or risk of institutional misconduct (Wilson et al., 2016).
•Both criminal history and the needs are captured with the tool. There is also an override feature to allow for the exercising of professional judgment to be exercised (Wilson et al., 2016).
Age Appropriateness
•Thirty four items are subdivided across ten subsections. The total score is used to calculate recidivism risk, categorised as either ‘minimum,’ ‘medium’ or ‘maximum.’ Subscale scores are used to identify criminogenic needs (Watkins, 2011).
•The tool centres on the principles of risk, need and responsivity, maintaining that those who are at high risk of reoffending should receive higher intensity interventions, supervision and monitoring (Watkins, 2011).
Year (1995)
Assessors must possess advanced training, certification and experience in psychological assessment or a related discipline, or satisfactorily complete a training course certified by the publishers. Can be used by a large range of professionals including social work and probation services.
•The LS instruments are based on ‘General Personality and Cognitive Social Learning theory,’ which is a general theory of criminal conduct entrenched in social learning perspectives (Wormith and Bonta, 2018).
•Lowenkamp et al. (2004) moderate to high levels of agreement observed across all ten subsections ranging from 61.5% to 97.7%.
Validation History
b)International Research
•The LSI R is supported by and reflective of three primary sources of information: (1) prior literature on recidivism, (2) professional opinions of probation officers and (3) social learning theory of criminal behaviours (Andrews and Bonta, 1995: 1).
•Hollin and colleagues (2003) found a 90% agreement rate in a sample of males.
•The subscales reflect the main risk factors identified in the research literature (Andrews and Bonta, 2010).
•Palmer and Hollin (2007) inter rater agreement levels of 95% for females
RATED page updated: July 2019 © Risk Management Authority 2019
•Subject to a number of meta analyses (Olver et al., 2014)
•Wilson and Stevenson (2017) claimed that the semi structured interview component of the instrument is a helpful framework for treatment and supervision, since it addresses learning, behavioural and developmental issues.
a)UK Research
General Predictive Accuracy
•Andrews (1982) excellent inter rater reliability coefficients ranging between .80 to .99.
•Dahle (2006) found excellent inter rater reliability generating an ICC value of .93 in a sample of German individuals who had offended.
a)UK Research
•Raynor and Miles (2007) predictive accuracy ranging from 65.4% to 71.6%.
Empirical Grounding
•Persson et al. (2017) found that the inter rater reliability for the LSI R was excellent (ICC=.92).
Inter-Rater Reliability
b)International Research
•Dahle (2006) the LSI R achieved moderate accuracy in violence prediction over a 10 year period (AUC =.65) in a sample of Germans
•Campbell French and Gendreau (2009) the LSI R displayed one of the largest mean effect sizes in predicting violent recidivism (Z+ =.28).
•Duwe and Rocque (2016) administered the LSI R to 26, 000 prisoners in Minnesota for the time period of 2003 to 2011. The results gave an AUC of 0.628, providing moderate support for the LSI R’s ability to assess need.
•Raynor and Miles (2007) for females in England and Wales (n = 163) the LSI R mean score = 21.2, % correctly predicted = 65%.
RATED page updated: July 2019 © Risk Management Authority 2019
Applicability: Females
•A study in Australia found that the LSI R yielded an acceptable level of reliability, with internal consistency estimates in the range of 0.59 to 0.784 (Watkins, 2011).
•Hollin and Palmer (2006) found a moderate correlation between the LSI R composite score and reconviction status.
•In a study of 828 prisoners in Midwest of the United States, the LSI R was able to predict recidivism (Smith et al., 2014).
•A study by Lowenkamp et al. (2009) found moderate correlations between both re arrest (r = .36) and re incarceration rates (r = .33) and the LSI R composite score.
Validation History
•Raynor (2007) LSI R presented ability to discriminate between reconvicted individuals who received a fine and those serving community/probationary sentences.
a)UK Research
•Palmer and Hollin (2007) found that for female prisoners in England and Wales (n = 150) the LSI R mean score = 23.0. There were significant differences between male and female scores on seven subscales, but not in
•Manchak et al. (2008) the LSI R yielded an AUC value of .73 for both general and violent recidivism.
Applicability: Ethnic Minorities
None at present.
RATED page updated: July 2019 © Risk Management Authority 2019
b)International Research
•Hsu, Caputi and Byrne (2010) the LSI R demonstrated small correlations with recidivism in a sample of male and female Australian Indigenous individuals (rs = .12 and .16 respectively). Indigenous individuals were found to score consistently higher on every item of the LSI R.
b)International Research
the overall score. Scores significantly predicted reconviction and time to reconviction. The composite score correctly classified 74%, with 79.7% correct classification for those not reconvicted and 64.9% for those who were convicted
•An Australian study found that the correlations between criminal history items and recidivism rates decreased in magnitude and significance when the LSI R was applied to females. The author posited that the LSI R subscales may not be suitable for fully assessing the criminogenic needs of females who offend (Watkins, 2011).
•Manchak et al. (2008) the LSI R attained excellent predictive accuracy in relation to recidivism in a sample of female who offended (AUC = .77).
•Fass et al. (2008) inconsistent validity with ethnic minority groups. LSI R had better predictive accuracy with Caucasians (80.4%) and Hispanics (82.4%) than African Americans (43.4%).
•Hogg (2011) found the LSI to be gender neutral.
Validation History
•In a meta analysis by Smith and colleagues (2009), it was found that the LSI R demonstrated a correlation of r=.35 for recidivism in females
•Vose et al. (2008) the LSI R was found to be a valid predictor of recidivism in females, achieving a composite score of 71.4% accuracy.
•Schlager and Simourd (2007) few statistically significant correlations between LSI R composite scores and recidivism amongst ethnic minority groups.
a)UK Research
•A study by Watkins (2011) found that the discriminatory power on the LSI R were very low for those with Aboriginal/Torres Strait Islander status in a sample of Australian individuals
Validation History
Applicability: Mental Disorders
•Applying the LSI R to 95 clients within a mental health jail diversion program, Lowder et al. (2017) determined that the LSI R showed weak predictive validity for African Americans than Caucasian clients. Moreover, the risk estimate was found to under classify African Americans for the moderate risk category; whilst over classifying them for high risk.
•Research by Lowder and colleagues (2019) suggested that there was no racial bias in the LSI R. Analysis focused on 11792 probationers in Kansas (74.7% White and 25.3% Black). Risk classifications and total scores produced similar levels of predictive accuracy between the two groups.
RATED page updated: July 2019 © Risk Management Authority 2019
•Chenane et al. (2015) examined the predictive validity of the LSI R in 2778 male prisoners in the Midwest of the United States across White, Black and Hispanic ethnic groups. Results indicated that the LSI R was better suited to predicting institutional misconduct for White prisoners than the other two groups. It was suggested by the authors that the tool is modified to adhere to the risks and needs of Black and Hispanic prisoners.
•Ostermann and Salerno (2016) applied the LSI R to 9454 individuals in New Jersey to gauge its validity in predicting recidivism within a year of their release from prison. It was found that the LSI R displayed low capacity for distinguishing between recidivists and non recidivists when applied to Black males.
•A meta analysis of 32 articles and 12 data sets was undertaken to examine whether the LSI R was applicable to Aboriginal individuals. Results indicated that all of the Central Eight risk/need factors were predictive of general and violent recidivism for Aboriginal individuals. Some of the factors demonstrated significantly better predictive validity for non Aboriginal individuals: criminal history, alcohol/drug and antisocial pattern (Gutierrez et al., 2013).
•Fewer validation studies conducted with other populations such as ethnic minority groups and mentally disordered individuals.
RATED page updated: July 2019 © Risk Management Authority 2019
•The LSI R has the ability to create awareness of a number of static and dynamic risk factors pertinent to the individual’s general risk of recidivism. Information obtained through the LSI R can inform the level and focus of monitoring and supervision strategies.
Contribution to Risk Practice
•A study assessed 193 detainees who were undergoing a forensic psychiatric investigation in Stockholm. The predictive validity of the LSI R was medium, generating an AUC of .70 (Persson et al., 2017).
Other Considerations
•Requires refresher training experience and training in the LSI R can affect the reliability of the instrument (Lowenkamp et al., 2009).
•The tool is a quantitative survey of risk need factors that are supported by research, professional opinion and social learning theory on criminal behaviour. It is not a comprehensive measure of mitigating and aggravating risk factors related to risk practices for offending (Andrews and Bonta, 1995).
•The tool can aid on going evaluation of an individual’s risk of reoffending and their criminogenic needs.
•The LSI R should be completed using information obtained from interviews with the individual and other collateral sources of information.
•The score of the LSI R was found to correlate with the HCR 20V3 and the SAPROF at a considerable rate; although the correlations between the risk or protection categories were poorer (Persson et al., 2017).
None at present.
•Harris, Rice and Quinsey (1993) found large weighted correlations ranging between .43 and .53 between items in the LSI and violent recidivism in a male psychiatric sample. Recidivists also tended to attain significantly higher scores on the tool than non recidivists.
a)UK Research
b)International Research
Similar specifications as with its predecessor, the LSI R.
Strengths
Empirical Grounding
Inter Rater Reliability
RATED page updated: July 2019
Name of Tool Level of Service Inventory Revised: Screening Version (LSI R:SV)
•The LSI R:SV can assist in prioritising cases for further intervention including assessment.
•Normed on Canadian institutionalised and probation populations.
•The LSI R:SV is supported by and consistent with ‘…general personality, social psychological theory of criminal behaviour and the LSI R items are consistent with an empirical body of literature and theory…’ (Andrews and Bonta, 1998:1).
a)UK Research
•Ideal for use when a complete LSI R assessment may not be feasible, due to time constraints or insufficient staff resources. It is estimated to take between 10 and 15 minutes to administer.
Year 1995
Category General Risk Assessment (Validated)
© Risk Management Authority 2019
None at present.
•Walters (2011) an estimate of inter rater reliability from a random sample of 17 participants revealed an ICC of .71.
Author / Publisher Andrews and Bonta
•Similar categorisation of risk as observed in the LSI R. High composite scores may warrant further analysis from the full LSI R or LS/CMI assessment.
Age Appropriateness
•The LSI R:SV is an 8 item actuarial screening tool derived from the LSI R. It encompasses seven key risk factors: criminal history, criminal attitudes, criminal associates, personal/emotional, employment, family and substance abuse.
16+ Assessor Qualifications
b)International Research
Description
b)International Research
•Lowenkamp et al. (2009) the LSI R:SV was not able to discriminate across the female risk categories of low, moderate and high. They recommend further larger sample research with subpopulations such as women.
Applicability: Females
Validation History
None at present
•Using a selection of 25 cases, Livingston et al. (2015) found that the LSI:R SV had an ICC of 0.79.
a)UK Research
a)UK Research
Validation History
Applicability: Ethnic Minorities
No Empirical Evidence Available.
•Walters and Schlauch (2008) the LSI R:SV demonstrated moderate predictive accuracy in relation to recidivism in a male prison sample with AUCs for (1) official records of at least one officially reported incident (.63), (2) official records of at least one ‘severe’ incident (.62) and (3) self reported incidents (.69).
•Yessine and Bonta (2006) compared 256 flagged individuals with 97 high risk violent ones. High risk violent individuals were found to scored lower on the LSI SV. Examining the predictive accuracy of the 256 flagged individuals sample of the LSI SV resulted in statistically significant results for all types of recidivism bar sexual recidivism. The AUCs generated were 0.68 for any recidivism, 0.67 and 0.63 for violent and non violent recidivism respectively; sexual recidivism yielded an AUC of 0.53. The authors caution that the lack of predictive power in relation to sexual recidivism may be due to the relatively low base rate.
RATED page updated: July 2019 © Risk Management Authority 2019
b)International Research
None at present
General Predictive Accuracy
Validation History
•The tool is useful for a brief scan of the main risk factors.
•The effectiveness of the LSI R:SV for screening the offending population is based on preliminary and limited evidence (Lowenkamp et al., 2009).
Applicability: Mental Disorders
•Thomas et al. (2009) found the LSI R:SV composite score generated moderate accuracy in predicting recidivism in a sample of forensic psychiatric patients (AUC = .72).
Contribution to Risk Practice
•The LSI R:SV can aid the assessor in identifying some static and dynamic risk factors pertinent to the individual’s likelihood of reoffending.
•Livingston and colleagues (2015) conducted a retrospective review of health records for 250 probationers with mental disorders. Predictive accuracy using the LSI R:SV was better for criminal justice contact and violent behaviour with AUCs of .61 and .67 respectively. The predictive power was less for non compliance and psychiatric adverse event with AUCs of .58 and .55 respectively.
•The LSI R:SV was applied to patients in a forensic psychiatric hospital in Australia to determine its scope to measure aggression risk. The results only showed a weak association between total scores and inpatient aggression, indicating that clinical factors pertaining to aggression should be incorporated into decision making (Daffern et al., 2005).
b)International Research
a)UK Research
Validation History
RATED page updated: July 2019 © Risk Management Authority 2019
•The tool can alert assessors to the need to conduct a more thorough assessment.
None at present.
Other Considerations
•Some research has found that the LSI R:SV does not discriminate between those at moderate and high risk (Lowenkamp et al., 2009).
•In a sample of 208 mentally ill individuals, the LSI R:SV predicted recidivism with moderate accuracy (AUC) for the following; (1) any new offence (.67), (2) for non violent new offences (.65) and (3) for violent new offences (.60) (Ferguson et al., 2009)
RATED page updated: July 2019 © Risk Management Authority 2019
•The LSI R:SV should be completed using information obtained from interviews with the individual and other collateral sources of information.
•Assessors should note that this tool is a screening version of the full assessments (i.e. LSI R, LS/CMI) and is not a comprehensive measure of risk and need factors.
18+
Strengths
•The OGP and the OVP predict “the likelihood of nonviolent and violent proven reoffending respectively” by combining information on identified static and dynamic risk factors (Howard, 2011: i).
•OASys includes a section dedicated to assessing the suitability of interventions.
OASys assessments must be completed by prison or probation staff who possess the necessary knowledge of behaviours of those who offend Continued refresher training on the administration and scoring of this tool is recommended.
•OASys also incorporates a self assessment component that allows the individual to record their views on their own risk/needs.
Age Appropriateness
Empirical Grounding
Assessor Qualifications
•OASys is grounded in the ‘what works’ evidence base as per risk need responsivity principles with regards to reducing reoffending (Moore and Howard 2015).
•OASys is an actuarial risk and needs assessment tool used by the prison and probation services in England and Wales.
RATED page updated: July 2019 © Risk Management Authority 2019
•The OASys is composed of 14 subsections and generates a summary risk score in order to assess likelihood of reoffending and risk of harm to self and others.
Name of Tool Offender Assessment System (OASys)
Description
Author / Publisher Home Office Year 2002
•In August 2009, the OASys General reoffending Predictor (OGP) and the OASys Violence Predictor (OVP) were introduced, and the old OASys score was discontinued. The August 2009 update also introduced ‘layered OASys,’ with Basic, Standard and Full assessments of similar structure but different length becoming available (Howard, personal communication, January 2013).
•An electronic version (eOASys) was introduced in 2005.
Category General Risk Assessment (Validated)
•The manual states that the measure’s development was founded from prison and probation effective practice guidelines and from empirical grounding of the LSI R and the Assessment Case management and Evaluation (ACE) (Home Office, 1999).
Validation History
a)UK Research
Validation History
a)UK Research
•Howard (2009) the OASys achieved moderate predictive accuracy. Accuracy of the instrument improved when used with the OVP and the OGP. The AUC values improved to 80% for non violent offending and 76% for violent re offending compared to 76% and 68% obtained from the OASys scores alone.
•Howard and Dixon (2012a) found that changes in OVP scores between the initial and final assessments, significantly predicted re offending in a sample of Welsh individuals.
Inter Rater Reliability
None available at present.
•Howard et al. (2006) 26% of persons rated as ‘low likelihood of reconviction’ were reconvicted within 24 months, compared with 58% assessed as ‘medium risk’ and 87% assessed as ‘high risk’.
RATED page updated: July 2019 © Risk Management Authority 2019
None available at present.
•Morton (2009) and Debidin (2009) most reliable items in the OASys were: accommodation, lifestyle/associates, drug misuse, Education, Training and Employability, Relationships, Emotional Well being and Attitudes. The least reliable items were: Financial Management, Alcohol, Thinking and Behaviour and Risk of Serious Harm.
•Debidin (2009) the OASys achieved moderate to high AUC values for different types of offending ranging from ‘Homicide and assault’ (.66) to ‘Weapons Possession’ (.74).
•Debidin (2009) moderate inter rater reliability was found in three case studies (ICCs ranged from .56 to .65).
b) International Research
b)International Research
General Predictive Accuracy
•Many of the factors identified by the OASys can act as targets for treatment/change.
None available at present.
•The OASys has the ability to create awareness of static and dynamic risk factors related to the individual’s risk of recidivism. It can also prompt further assessment of identified risk factors.
No Empirical Evidence Available
a)UK Research
Howard and Dixon (2012a) report: "The present psychiatric treatment item seems crude but reliable: unlike other items in section 10 (Emotional Well being), only basic information and training are required to score it. Most OASys assessments record little or no direct information on personality disorder, psychopathic personality features or active psychotic symptoms."
•Predictive validity for black and ethnic minority groups was found to be lower in a study by Howard (2015a)
Contribution to Risk Practice
Validation History
a)UK Research
•The OASys assessments, including the OGP and OVP scores, are summarised to inform Pre Sentence Reports within the National Probation Service for England and Wales.
•Factors included in the OASys can inform offence analyses and risk formulations.
b)International Research
•Debidin (2009) the OASys attained moderate to high AUC values of .72 to .81 for violent and non violent offences in a female offending sample (n=1,585)
Applicability: Ethnic Minorities
•Debidin (2009) low to high AUC values obtained for individuals of other ethnic minorities ranging from .57 to .75 for violent and non violent offences.
•Howard and Dixon’s (2012a) study of the OVP recorded 7% non white participants in the 2002/2004 cohort and 8% non white participant representation in the 2004/2005 data set.
None available at present.
Validation History
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
Applicability: Females
Applicability: Mental Disorders
•Based on a validation study of almost fifteen thousand individuals who had committed sexual offences, a sexual offending component the ‘Sexual Predictor’ has been added to the OASys in order to predict contact sexual reoffending (Howard and Barnett, 2015).
•The National Offender Management Service reserves Crown Copyright on OASys. Please contact mark.nickson@noms.gsi.gov.uk for details on licensing.
•Fitzgibbon and Green (2006) and Fitzgibbon (2008) concerns relating to the accuracy of the OASys in predicting recidivism in sub groups of such as those with mental disorders and ethnic minorities. Other concerns regarding its utility in aiding parole decisions.
•The compendium will also include a study constructing and validating a predictor of sexual reoffending, provisionally named the OASys Sexual reoffending Predictor (OSP) (Howard, personal communication, January 2013).
•Howard (2015c) had a series of recommendations with regards to the positive factors of the OASys: assessors should recognise the importance of both positive (personal strengths) and risk factors, something which could be highlighted during training; monitoring the recording of positive factors to ensure current ones are being maintained and to be aware of the development of others.
•Few validation studies published in peer reviewed journals. Majority of validation studies conducted by the Home Office.
RATED page updated: July 2019 © Risk Management Authority 2019
•The OASys can contribute to risk management plans for more complex cases which require intensive monitoring and more detailed offence analyses. The tool contains a ‘Risk of Serious Harm’ section which allows the assessor to identify factors related to this construct.
•The implementation date of OGP2, OVP2 and OSP has not been confirmed.
•Morton (2009) and Debidin (2009) limited inter rater reliability of some of the subsections on the OASys.
Other Considerations
•OGP Version 2 (OGP2) and OVP Version 2 (OVP2) have been peer reviewed and will be published in a forthcoming Ministry of Justice publication (Howard, in preparation). The OGP2 and OVP2 includes an ‘offence free time’ component, which enables estimates to be made for those who have spent time in the community without reoffending, given that the likelihood of reoffending is greatest immediately after sentence (Howard, 2011). The compendium also includes validation evidence for the predictive accuracy of the OGP2 and the OVP2 in different offending groups according to age, gender and ethnicity.
•The tool is used in conjunction with the OASys: this is designed to assess how likely an individual is to reoffend, identify and classify offending related needs (Moore, 2015). It can also be used in cases where the OASys has not been completed.
Name of Tool Offender Group Re Conviction Scale Version 3 (OGRS3)
Description
•Due to a reduction in the number of items from nine to six, OGRS3 can be scored more quickly and accurately than previous versions (Howard, 2018).
•It can provide a prediction of risk within a 1 to 2 year time period
18+ Assessor Qualifications
Author / Publisher Howard and colleagues
Strengths
Age Appropriateness
•It provides a gendered estimate of risk, calculating it differently for females and males (Howard et al., 2009; Howard, 2018).
•The OGRS3 is an actuarial assessment tool that is used in conjunction with the OASys risk assessment by the National Offender Management Service (NOMS) (Home Office, 2002) in order to inform and improve the static/dynamic predictor found in the OASys (Howard et al., 2009). It was originally owned by the Home Office and was later transferred to Her Majesty’s Prison and Probation Service in England and Wales (Howard, 2018).
•A fourth version of the tool has been developed and is currently awaiting release. This includes a predictor of non sexual, violent recidivism (OGRS4/V) (Howard, 2018).
Empirical Grounding
Category General Risk Assessment (Validated)
RATED page updated: July 2019 © Risk Management Authority 2019
Year 2009
•The OGRS3 contains items pertaining to the age at time of current caution, the type of offence, prior criminal history (including duration in years) and gender of the individual being assessed (Stephens and Brown, 2001).
Qualified Probation Officer with the relevant training and experience.
•The tool generates a probability of reconviction (Stephens and Brown, 2001).
•Wood et al. (2015) noted a link between reoffending rates and the OGRS score. Thirteen percent of those with a ‘very low’ likelihood of reoffending went on to do so in comparison with the 67% of those ranked as ‘very high’ who recidivated.
•The developers explored previous research in the United Kingdom about gender, age, current offence and criminal history as significant predictors for recidivism. Subsequent versions were refined by testing the validity across different groups of individuals who have offended; this then led to the age/gender interaction in OGRS3 (Howard, 2018).
Inter Rater Reliability
No Empirical Evidence Available.
Validation History
•The criminal history ‘copas rate’ is the most complex part of the OGRS based on two factors: the length in years of an individual’s known criminal career and their total number of convictions. The ‘copas rate’ of an individual is higher when they have more criminal appearances within a short ‘criminal career’ (i.e. from their first through to their current offending) (Howard, 2018).
General Predictive Accuracy
•The OGRS3 is grounded in extensive Home Office Policy research (Kershaw, 1999, Independent Conference Paper) dating from the previous two versions of the tool (Copas and Marshall, 1998; Taylor, 1999).
a)UK Research
•Howard and Dixon (2012b) the OGRS3 attained moderate accuracy in predicting violent reoffending (AUC = .70) in a dataset of 49,346 assessments.
•Wakeling et al. (2011a) significant differences in mean OGRS3 scores between the recidivist and non recidivist groups (20.3 versus 9.6 respectively). OGRS3 obtained moderate to high AUC values with different groups ranging from .65 (those convicted of sexual offences) to .86 (those convicted of violent offences).
•Howard et al. (2009) the OGRS3 substantially improved the prediction of ‘proven’ re offending for all individuals (AUC = .80), compared with its predecessor, the OGRS 2 (AUC= .78). For prisoners only, the OGRS3 generated an AUC of .84 compared to the Sentence Planning Predictor (AUC <.83) (n = 71, 914).
RATED page updated: July 2019 © Risk Management Authority 2019
•Howard (2018) found that the AUC was strong at .80 for when coding was carried out using centrally held records
•The OGRS3 has an ability to guide awareness of some static risk factors and can prompt further need for assessment of the risk of reoffending.
b)International Research
•Debidin (2009) the OGRS3 obtained moderate to high AUC values of .81 and .70 for non violent and violent offending respectively in a female offending sample.
Applicability: Females
RATED page updated: July 2019
b)International Research
Applicability: Mental Disorders
© Risk Management Authority 2019
Validation History
None available at present.
a)UK Research
•Debidin (2009) the OGRS3 obtained moderate to high AUC values ranging from .64 for those convicted of violent offences who were of ‘Other’ ethnic origin to .75 for non violent mixed race individuals.
None available at present.
•Howard et al. (2009) maintained that the predictive performance of OGRS3 can be optimised by using the OASys or the Asset (Youth Justice Board, 2003) in conjunction with it.
No Empirical Evidence Available.
•Howard (2009) reported OGRS3 provided more accurate predictions for females than the previous version.
Validation History
b)International Research
•OGRS3 scores are used within the National Offender Management Service (NOMS) as part of the risk/needs/responsivity based criteria for targeting of offending behaviour programmes.
and when the coding was completed by probation workers.
Applicability: Ethnic Minorities
a)UK Research
Contribution to Risk Practice
None available at present.
Validation History
•Assessors should note that the OGRS3 is designed to be used in conjunction with the OASys; hence the observed limitations in its capacity to contribute to risk practices on its own.
•Few validation studies by independent researchers.
•OGRS prediction scales are used as a base measure in a number of settings. For example, the Ministry of Justice is linking OGRS with the 'payment by results' scheme. In Wales OGRS was used to evaluate a mentoring scheme for ex prisoners. (Maguire et al., 2010).
•The Youth Justice Board is in the process of rolling out the use of version 4 of the OGRS for young people (Moore and Howard, 2015).
•A note of caution is potentially an individual’s score can fall when they receive a new conviction. This is in scenarios when the length of the ‘criminal career’ is longer than the number of convictions or when age increases and an individual goes up in an age band. The decrease in OGRS score is to reflect the effect of growing older and longer breaks between offences. This should be considered, however, in the context of the behaviours and circumstances of individuals (Howard, 2018).
•Authors claim that OGRS3 can be used within the youth justice system; although there is no empirical evidence to date to support this claim.
Other Considerations
RATED page updated: July 2019 © Risk Management Authority 2019
•OGRS3 does not have a component to capture violent recidivism. The fourth version of the tool addressed this gap (Howard, 2018).
•Howard (2015a) found that the OGRS3 accurately measured rarer types of serious offences. It was recommended that arson, kidnapping, blackmail, dangerous driving and racially aggravated offences should be added to the OGRS3.
•OGRS Version 4 (OGRS4) will be introduced in a forthcoming Ministry of Justice publication (Howard, in preparation). The publication date has not yet been confirmed. Preliminary research has found the OGRS4, consisting of general and violent reoffending models, significantly outperforms the third version (Howard, 2015b).
•Formal training for the OGRS is provided for prison staff members as part of the OASys assessor course (Howard, 2018).
•The OGRS4 includes a separate predictor of violent recidivism as well as a predictor of general recidivism. Both remain based on static risk factors, though OGRS4 also includes violent offending history. The OGRS4 publication includes tests of validity by age and gender, including young people OGRS4 includes an ‘offence free time’ component, which enables estimates to be made for those who have spent time in the community without reoffending, given that the likelihood of reoffending is greatest immediately after sentence (Howard, 2011).
•The National Offender Management Service reserves Crown Copyright on OGRS versions 3 and 4. Please contact mark.nickson@noms.gsi.gov.uk for details on licensing.
•Miller (2006) found that those who served longer custodial sentences and had more arrests for non violent offences had higher composite scores in the IORNS than those who served shorter prison sentences and had fewer arrests for non violent offences.
Category General Risk Assessment (Awaiting Validation)
Men: 18 75; Women: 18 60
•The IORNS is a self report actuarial measure used for the assessment of static risk, dynamic needs and protective strength factors in relation to reoffending, treatment needs and management in adults (Miller, 2018).
Author / Publisher Miller Year 2006
Name of Tool Inventory of Offender Risk, Needs and Strengths (IORNS)
Description
•A study by Miller (2015) found that the likelihood of reoffending increased when those who had committed sexual offences increased their favourable impression on the IORNS. The conclusion was reached that self perceived protective strengths were significantly predictive of recidivism for general, sexual and violent offending.
Can be administered by persons who do not have training in forensic or clinical psychology or psychiatry, with supervision and interpretation by a licensed or certified professional.
Tool Development
Assessor Qualifications
•Bergeron and Miller (2013) found that the measurement properties of the dynamic needs index (DNI) are acceptably invariant over time. There was also evidence that the intercept of alcohol/drug problems scale is higher before treatment and the intra/interpersonal problems scales are higher before treatment.
General Notes
Age Appropriateness
RATED page updated: July 2019 © Risk Management Authority 2019
•There are 130 items in total providing 4 indexes: a static risk index (SRI) of historical (unchanging) items related to offending behaviour and recidivism; dynamic need index (DNI), targeting specific areas of risk related to offending behaviour; protective strength index (PSI), examining factors to promote resilience. All of these are calculated to produce a composite score in the form of the overall risk index (ORI) (Miller, 2018; Ullrich and Coid, 2011).
•Since it is a self reported measure, the scores of needs, risk and strengths are all the individual’s own perception (Miller, 2015).
•The creator cautions that the IORNS is not meant to be an actuarial tool (Miller, 2018).
•Treatment providers and evaluators can infer from the IORNS scores possible hypotheses around problem areas, needs and treatment progress (Miller, 2018).
RATED page updated: July 2019 © Risk Management Authority 2019
•The IORNS includes two validity scales: the ‘Inconsistent response style’ (IRS), checking for consistency between answers; the ‘favourable impression scale’ to determine whether the individual was trying to portray themselves positively (Miller, 2018).
Category General Risk Assessment (Awaiting Validation)
Author / Publisher Fazel, Change, Fanshawe, Långström, Lichtenstein, Larsson and Mallett
•Socioeconomic deprivation is defined via a standardised, normalised score, including rates of welfare recipiency, unemployment, poor education, crime rates and median income in an individual’s residential area (Fazel et al., 2016).
•Violent offences within this tool refer to homicide, assault, robbery, arson, any sexual offence (rape, sexual coercion, child molestation, indecent exposure or sexual harassment), illegal threats or intimidation (Fazel et al., 2016).
Assessor Qualifications
Tool Development
•The variables considered for inclusion were drawn from the existing evidence of criminal history and sociodemographic and clinical factors (Chang et al., 2015; Fazel et al., 2012).
Year 2016
•A fourteen item tool was derived using Swedish population registers (sample size=37,100) and externally validated on a sample of 10,226 individuals. Risk of violent reoffending at the 1 year time point indicated a sensitivity of 76% and a specificity of 61%. At 2 years, the sensitivity and specificity were 67% and 70% respectively. The external validation model displayed good discrimination for violent reoffending within 1 year (AUC=0.75) and 2 years (AUC=0.76) after prison release. Good
•This tool is designed to predict violent reoffending in individuals being released from prison after 1 or 2 years (Fazel et al., 2016).
Name of Tool Oxford Risk of Recidivism Tool (OxRec)
•Fourteen variables are included in the tool: gender, age, immigrant status, length of incarceration, violent index offence, previous violent offence (before index offence), neighbourhood deprivation, income level, mental disorders, civil status, highest education, employment, disposable income, neighbour deprivation scale, alcohol abuse, drug abuse, any mental disorder, any severe mental disorder (Fazel et al., 2016).
Although no specific training or qualifications are required to use the tool, appropriate application and scoring of cases requires the judgment of criminal justice or healthcare professionals.
Age Appropriateness
RATED page updated: July 2019
© Risk Management Authority 2019
Description
•The tool categorises individuals into three level of risk: low, <10% risk; medium, 10 50% risk; high, >50%. If one or more of the variables are set to ‘unknown,’ then a range of risk levels are displayed (Fazel et al., 2016).
16+
•OxRec has recently been validated in a national sample of individuals who have offended in the Netherlands; although the model required recalibration prior to use. This showed moderate discrimination with an AUC of 0.68 for 2 year violent reoffending and 0.69 for any reoffending in the prison cohort. Adequate calibration scores were also shown (Fazel et al., 2019).
•OxRec takes around 10 to 15 minutes to complete, relies on mostly routinely collected information, is freely available and does not require any formal training (Fazel et al., 2019).
RATED page updated: July 2019 © Risk Management Authority 2019
•Some items of the OxRec are not easily generalizable to other countries and may require modification (e.g. neighbourhood deprivation score) (Fazel et al., 2019).
•The authors indicate that criminal justice, forensic and healthcare professionals might take different approaches to using such a tool. Prison healthcare may use it to treat prisoners before their release or by probation services or case workers to plan sentencing and release arrangements (Fazel et al., 2016).
•This tool has only been validated in Sweden and the Netherlands thus far (Fazel et al., 2016; Fazel et al., 2019). Other validations are in progress.
calibration was also evident for violent reoffending at 1 and 2 years after prison release, with Brier scores of 0.095 and 0.108 respectively (Fazel et al., 2016).
•The tool is freely available online: https://oxrisk.com/oxrec/
•In terms of timing, OxRec could be used towards the end of prison sentences to assist with post release management of risk, including linkage with community addition and mental health services (Fazel et al., 2016).
General Notes
•OxRec is available in English, Swedish, Greek, French and Chinese versions.
•The SAQ is a 72 item actuarial self report assessment consisting of true and false questions. The purpose of it is to predict violent and non violent recidivism among adults who have offended (Mitchell, Caudy and Layton, 2012).
Age Appropriateness
•These 72 items are spread across seven subscales included in the assessment: (1) Criminal Tendencies (antisocial attitudes, beliefs, feelings and behaviours); (2) Antisocial Personality Problems, looking at characteristics similar to those covered in antisocial personality disorder; (3) Conduct Problems (assesses childhood behavioural issues); (4) Offender’s criminal history; (5) Alcohol and drug abuse; (6) Antisocial Associates. These six subscales are used to predict recidivism. There is an Anger subscale, measuring reactions to anger; however, this is not included in the total score because of the controversial relationship between anger and recidivism. This scale is instead used to assign individuals to treatment programmes dealing with anger. Also included is a Validity subscale for validating an individual’s truthfulness in responding to SAQ Items (Loza, 2018).
Author / Publisher Loza Year 1996
•Of the 72 items presented in the assessment, only 62 items are used to predict recidivism. The remainder of statements may assist with determining issues such as substance abuse and personality disorders. Risk is classified as ‘low’, ‘low moderate,’ ‘moderate,’ ‘high moderate’ and ‘high (Loza, 2018).
•The SAQ was initially developed to cover the main themes found in the recidivism literature (most prominently featured in anti social theories) (Loza, 1996).
Category General Risk Assessment (Awaiting Validation)
RATED page updated: July 2019 © Risk Management Authority 2019
Tool Development
18+
Assessor Qualifications
The SAQ can be administered by a variety of forensic professionals: psychologists, psychiatrists, parole officers, behavioural technologists, nurses and others trained in administering psychological tests or questionnaires. A minimum of The assessor should have graduate level training and qualifications in administering other similar tests and measures. (Loza, 2018).
Description
Name of Tool Self Appraisal Questionnaire (SAQ)
•The SAQ could potentially be used to determine the most appropriate treatment program, e.g. if an individual who has offended has a high score on the SAQ anger sub scale, an anger management program could be offered to them (Loza, 2018).
RATED page updated: July 2019 © Risk Management Authority 2019
•Villeneuve, Oliver and Loza (2003) found the composite SAQ scores were significantly higher in the high risk psychiatric sample when compared to general correctional individuals with no history of major psychiatric illness. Moderate to large correlations ranging from .28 to .50 were found for violent recidivism, general recidivism, ‘new sentence’ and ‘any failure’.
•Loza et al. (2005) found moderate predictive accuracy (AUC =.70) in relation to re incarceration in a sample of females. No significant differences were found between the responses of African American individuals compared to Caucasian ones.
•Majority of research has been conducted or co authored by the author of the SAQ assessment; although some international studies have emerged in recent years.
•Mitchell and Mackenzie (2006) found that the SAQ was unable to predict recidivism in high risk individuals with drug offences. However the findings of this study have been disputed by the author of the SAQ on the grounds of limitations in the methodology, statistical analyses and sample selection (see Dhaliwal, Loza and Reddon, 2007).
•Kubiak et al. (2014) assessed the usefulness of the SAQ with a sample of 534 incarcerated females. Whilst self reported violence was considered to be a strong predictor of SAQ scores, many of the women in the most violent group did not reach the cut off points in their scoring. To that end, the authors suggest that the scoring thresholds are modified for females in order to adequately assess their treatment needs.
General Notes
•Loza, Loza Fanous and Heseltine (2007) in a 9 year follow up study, the SAQ demonstrated a sensitivity of 59% for non violent recidivism and 70% for violent recidivism. It was also found that the SAQ had a specificity of 74% for non violent recidivism and 62% for violent recidivism.
•A study of 125 males in South Africa found that the SAQ produced reliable scores, suggesting it is appropriate for application in this country (Prinsloo and Hesselink, 2011).
•Andreau Rodriguez, Peña Fernández, and Loza (2016) administered the SAQ to 276 individuals in Spain to test its ability to measure recidivism. Recidivism in this study was defined as a second or subsequent entry in prison by the same person for committing a violent crime in the community. The SAQ showing acceptable accuracy in discriminating between violent and non violent recidivists, with the total score generating an AUC of .80.
•Hemmati (2004) in a sample of individuals aged between 12 and 20 years, significant differences were found between violent and non violent individuals on the SAQ total scores.
•Mitchell, Caudy and Layton (2012) found that the SAQ total score yielded a modest prediction of reconviction, accurately predicting this for circa 63% of all possible pairs of individuals.
•Rodrigues and colleagues (2016) applied the SAQ to 121 males within a correctional facility for mental health issues. It was found that the tool significantly predicted general recidivism (AUC=.74) and predicted institutional aggression (includes threats, verbal aggression or assault) (AUC=.61).
•The SAQ showed moderate correlations with the PCL R and VRAG total scores, suggesting a degree of concurrent validity (assessing measures to see if they produce similar results) (Andreau Rodriguez, Peña Fernández, and Loza, 2016)
•For the most accurate predictions, Mitchell, Caudy and Layton (2012) advised using the SAQ total scores with those generated by age and number of prior arrests.
Age Appropriateness
No assessor qualifications specified at present
•The Women’s Risk/Needs Assessment is designed to assess both gender neutral (e.g. criminal history) and gender responsive (e.g. self efficacy) factors in females who have offended (Van Voorhis, Bauman and Brushett, 2012). The gender responsive factors are: relationship support and conflict, parental involvement and stress, self efficacy, prior physical and sexual trauma, housing, safety, mental health and anger/hostility. The gender neutral items are: past and current substance abuse, criminal history, employment and financial stability, educational strengths and needs and antisocial attitudes (Beppre and Salisbury, 2016).
Category General Risk Assessment (Awaiting Validation)
18+
RATED page updated: June 2019 © Risk Management Authority 2019
Author / Publisher National Institute of Corrections/University of Cincinnati (NIC/UC)
•As part of the assessment, individuals are interviewed. After that, they complete a self report survey assessing additional gender responsive factors (Brushett, 2013).
Tool Development
Description
The University of Cincinnati Corrections Institute offers a 3 day training course in the administration of the WRNA and a 1 day booster course for those who have already the training which is customised for each individual site through survey feedback. Training courses are also offered in training the trainers, agency wide training and quality assurance. Web based individualised orientation and consulting sessions are available to agencies interested in learning more about the adoption of the WRNA.
Name of Tool Women’s Risk/Needs Assessment
Assessor Qualifications
•The current assessments include; (1) the full instrument, Women’s Risk/Needs Assessment which contains separate forms for pre release, probation, prison settings and (2) the Women’s Supplemental Risk/Needs Assessment (‘Trailer’) which is designed to supplement existing gender neutral risk/need assessments such as the Level of Service Inventory (LSI) (Van Voorhis, Bauman and Brushett, 2013).
Year 2011
•The Trailer is not a screening version of the full assessment, rather, it is solely comprised of the gender responsive factors contained within the full assessment which is used to supplement other validated risk assessment tools (Van Voorhis, Bauman and Brushett, 2013).
•The tool forms a response to the issues raised within the literature in terms of the gender specific factors that increase the likelihood of offending in females: histories of victimisation and abuse, relationship problems, mental health issues, substance abuse, self efficacy/confidence, poverty and parental stress (Van Voorhis et al., 2010).
•The WRNA T improved the predictive validity of the LSI R by providing a means for screening the gender responsive needs documented in the LSI R. AUCs ranged from .55 to .77 for six months and .79 for 12 months for the LSI R on its own. When the WRNA T was added, predictive validity improved from .55 to .77 for outcomes at six months and .59 to .80 for twelve month outcomes (Van Voorhis, Bauman and Bruschett, 2013).
•Wright, Salisbury and Van Voorhis (2007) low to moderate associations found (correlations coefficients ranging from r=0.9 to r=.20) between the composite and subscale and item scores with institutional misconduct amongst incarcerated females. Moreover, their results showed that gender responsive needs (r=.27 to r=.34) in some cases performed slightly better than gender neutral ones (r=.23 to r=.33) when predicting institutional misbehaviour.
General Notes
•The aim of this tool is to provide a structured assessment that will identify and link women to meaningful programs and services (Van Voorhis et al., 2010).
•van Voorhis, Bauman and Bruschett (2013) found that the use of the WRNA T to supplement the LSI R in Rhode Island made it a stronger predictor of risk.
RATED page updated: June 2019 © Risk Management Authority 2019
•Preliminary studies have investigated the construction and validation of the items presented within the tool (Van Voorhis et al., 2010; Wright, Salisbury and Van Voorhis, 2007).
•Current use of the assessment requires a written agreement with the University of Cincinnati’s Office of Intellectual Property (Bauman, personal communication, February 2012).
•For further information, please visiT http://www.uc.edu/womenoffenders.html or e mail enquiries to Ashley Bauman (ashley.bauman@uc.edu) or John Schwartz (john.schwartz@uc.edu).
•The assessment is based on prior literature relating to the trajectories of offending in female populations (Wright, Salisbury and Van Voorhis, 2007; Van Voorhis et al., 2010).
•Many of the items contained in the assessments were developed by the members of the Women’s Issues Committee of the Missouri Department of Corrections in collaboration with researchers at the University of Cincinnati (Van Voorhis et al., 2008).
•The gender neutral factors are based on existing risk assessments such as the LSI.
RATED page updated: July 2019 © Risk Management Authority 2019
•Can be used with females, aboriginal, psychopathic and mentally disordered individuals.
Assessors are required to undertake an intensive training course.
Category Violence Risk (Validated)
Description
The VRS can be considered as a ‘conceptual actuarial scale,’ since the risk predictors are primarily derived from the Psychology of Criminal Conduct, a text by Andrews and Bonta (2010) that utilises personality, cognitive behavioural and social learning perspectives to conceptualize the psychology of criminal behaviour Its static and dynamic risk factors are empirically and/or theoretically related to violent recidivism (Wong and Gordon, 2006).
Strengths
•The VRS is a 26 item actuarial risk assessment tool designed to assess the risk of violent re offending for incarcerated individuals and forensic psychiatric patients being considered for community access.
•The tool consists of six static and twenty dynamic variables. It can be used to monitor variations in risk and motivation to change. The second edition includes an item ‘criminal personality’ intended to capture the characteristics of psychopathic individuals (Dolan et al., 2008).
•Assesses risk changes as a function of treatment or variations over time.
Empirical Grounding
Assessor Qualifications
Can be used by workers within the criminal justice system. No professional qualifications required.
•Assess risk of violence using a combination of static and dynamic (changeable) risk factors, the latter can be used to identify treatment targets.
18+
•A discretionary clinical over ride is available for situations that are not captured by the risk factors found in the tool.
Author / Publisher Wong and Gordon Year 2001
Name of Tool Violence Risk Scale Second Version (VRS 2)
•Assesses treatment readiness/motivation which can inform approaches to treatment.
Age Appropriateness
•In an unpublished PhD thesis, Stewart (2011) looked at the VRS ratings of 101 federally sentence women in Canada were followed up for approximately 7 years in the community. ICC was .98, AUCs for violent recidivism and institutional misconduct were .84 and .78 respectively.
a)UK Research
b)International Research
•Lewis, Olver and Wong (2013) reported ICC values ranging between .82 to .84 for the VRS total score in a sample of high risk male Canadian individuals who offended with significant psychopathic traits.
Validation History
The VRS has been developed for use in criminal justice and forensic psychiatric settings (see section ‘Applicability: Mentally Disorders).
General Predictive Accuracy
•Zhang et al. (2012) reported an ICC of .80 for the VRS total score in a sample of male and female Chinese forensic inpatients in the province of Sichuan, all of whom were suffering from significant mental disorders.
Validation History
a)UK Research
Applicability: Females
•Dolan et al. (2008) reanalysed their results removing the data of female participants from the sample; however, this did not significantly alter the previous findings. It should be taken into consideration that the female sample size was very small (n=11).
b)International Research
•Wong and Parhar (2011) found an ICC value of .93 for the VRS total score in a sample of Canadian males on parole or other forms of conditional release in the community.
•Doyle et al. (2012) the VRS total score attained an ICC value of .96 in a sample of male and female patients discharged from acute mental health units.
•Dolan et al. (2008) reported high correlation coefficients for inter rater reliability of the VRS composite score, static subscale and dynamic subscale (ICCs = .89, .96 and .85 respectively).
RATED page updated: July 2019 © Risk Management Authority 2019
Inter Rater Reliability
Applicability: Ethnic Minorities
RATED page updated: July 2019 © Risk Management Authority 2019
•Dolan and Fullam (2007) the VRS was able to discriminate violent and non violent patients As an effect size test used to indicate the standardised difference between two means, Cohen’s d was equal to .72. Patients who had engaged in institutional violence in the following 12 months post assessment had higher mean VRS composite and subscale scores than the non violent group.
•Wong and Parhar (2011) reported AUC values of .83 and .72 in predicting violent and any re offence respectively after 7 years of prospective follow up in the community
Contribution to Risk Practice
•Wong and Gordon (2006) the VRS had attained high AUC values in predicting recidivism in the following domains: ‘all convictions’ (AUC = .74), ‘violent convictions’ (AUC = .75), and ‘non violent convictions’ (AUC = .72).
•Lewis, Olver and Wong (2013) reported positive results using the instrument with high risk individuals with psychopathic traits. In a fixed 3 year follow up period (n=110), the VRS post treatment total score was predictive of violent reconvictions (AUC =.65); however, the pre treatment total score was not significant (AUC=.60); with a variable follow up period (n=150), both pre and post treatment total scores were significant (AUC=.60, .64 respectively).
The normative sample consists of approximately 45% aboriginal males (Wong & Gordon, 2006).
Validation History
•Dolan et al. (2008) reported moderate to high predictive accuracy of the VRS 2 with the occurrence of an aggressive incident in relation to the composite score, the static subscale score and the dynamic subscale score (AUCs = .69, .60 and .70 respectively). The authors tentatively recommend the use of VRS 2 to predict inpatient violence.
a)UK Research
b)International Research
Validation History
Applicability: Mental Disorders
RATED page updated: July 2019 © Risk Management Authority 2019
•The VRS consists of 20 dynamic factors that can be used to assess risk and identify treatment targets, inform the formulation of risk management plans and in release decision making
Other Considerations
•The second edition (VRS 2) was an experimental version so named when it was under development. The content of the VRS and VRS 2 are essentially the same with only minor changes. Currently, the VRS is the appropriate name for the tool.
•For training to use the tool clinically, and for additional research on the tool, please contact the authors (s.wong@sasktel.net or audrey.gordon@outlook.com).
•The VRS incorporates the Stages of Change model within the dynamic risk factors to assess treatment readiness and risk change. Using the combination of dynamic risk factors and the assessment of treatment readiness and change, a VRS assessment can also inform the levels of monitoring and rehabilitation efforts and risk change over time or with treatment (Wong and Gordon, 2006; Lewis et al., 2012; Olver et al., 2013).
•A screening version of the tool was developed to highlight which individuals may require more in depth assessments or to be used for brief intake evaluations (contact authors for more information, see below)
Strengths
•After the assessor has completed an interview with the participant, the software generates a report that consists of statistical estimates of the likelihood of future violence, including the confidence interval for that estimation of violence (Monahan, 2010).
•The COVR is a self report interactive software programme that aims to estimate the level of violence risk posed by individuals diagnosed with a mental disorder over a period of several months, post discharge into the community.
•A ‘classification tree’ methodology is used in the COVR, with questions being asked in a sequence until the individual is assigned a risk category (Monahan, 2010).
18 60
Assessor Qualifications
•The COVR provides an estimate of risk that the practitioner can consider in relation to any other information they hold about the patient (Monahan, 2010).
•The tool was developed from the MacArthur Violence Risk Study (Monahan et al., 2001). Variables from the study that predicted future violence were then used to classify participants into risk categories using the iterative Classification Tree System. This is an interactive model of violence,
•Software based assessment that can eliminate sources of error and has the practicality of screening large samples (Snowden, et al. 2009).
Assessors must possess the relevant qualifications (i.e. in the administration and interpretation of psychological tests) and training in the use of this tool.
Category Violence Risk (Validated)
•In 2007, McClusker expressed uncertainties about the instrument. Counteracting that, Meadows (2014) reported a study that "confirmed that COVR scores were predictive of re hospitalization or violent recidivism."
Name of Tool Classification of Violence Risk (COVR)
Empirical Grounding
Description
RATED page updated: July 2019 © Risk Management Authority 2019
•The tool assesses patients on 44 risk factors in estimating violence risk (Monahan, 2010).
Age Appropriateness
Author / Publisher Monahan and Colleagues
Year 2005
•Assessment is fairly quick to administer
b)International Research
•Snowden, Gray and Taylor (2010) found that applying the instrument in a UK sample indicated "COVR was a good predictor of both verbal and physical aggression. Its predictive ability was similar to that of the VRAG, although the VRAG was a better predictor of violence to property."
RATED page updated: July 2019 © Risk Management Authority 2019
None available at present
•The COVR implements actuarial methods to a long established modelling approach used in the medical field to inform professional judgements (Monahan et al., 2001).
Validation History
Validation History
a)UK Research
b)International Research
None available at present.
•Monahan (2010) Using a dataset of 385 interviews, high kappa coefficients were found.
Inter Rater Reliability
This tool can be applied to female adults who have offended; however, there is limited research validation relating to this population.
Applicability: Females
General Predictive Accuracy
No Empirical Evidence Available.
considering combinations of risk factors in order to classify an individual into a risk level (Monahan, 2010).
Applicability: Ethnic Minorities
Validation History
a)UK Research
•Meadows (2014) indicated: "The current study assessed success rates in predicting violent recidivism among forensic and civilly committed inpatients released from a state psychiatric hospital utilizing the violence risk categories generated from the COVR. This study confirmed that COVR scores were predictive of re hospitalization or violent recidivism."
•Persson et al. (2017) assessed 200 individuals (193 of which were followed up) who were undergoing forensic psychiatric investigation in Stockholm. The predictive validity for the COVR tool was found to be modest with an AUC of .61.
RATED page updated: July 2019 © Risk Management Authority 2019
a)UK Research
•Sturup, Kristiansson and Lindqvist (2011) in a 20 week follow up period post discharge, the COVR attained a high AUC value of .77 in predicting violent offences within a sample of 331 forensic mental health patients.
b)International Research
•Monahan et al. (2005) the COVR obtained an AUC value of .70 (sensitivity = .75 and specificity = .77) in their follow up investigation of patients from the MacArthur study.
•Issues regarding the reliability of self report (Snowden et al., 2009).
Validation History
•The tool provides a statistical probability of short term risk of recidivism (i.e. within a one year period following discharge from a secure facility). No other supplementary information is included
•Snowden et al. (2009) in a six month follow up the COVR presented the ability to predict physical aggression (AUC = .73); however, it was unable to predict property offences and verbal aggression.
•The tool is useful in aggregating the individual’s self reports in relation to the formulation of the risk of reoffending.
•Doyle et al. (2010) no significant findings were found in relation to the tool’s predictive accuracy and inpatient violence in a 20 week post discharge follow up.
•McDermott et al. (2011) COVR has modest predictive accuracy in relation to physical aggression by psychiatric patients (AUC = .73) in the 20 week follow up period subsequent to administration of the tool.
Applicability: Mental Disorders
•Monahan (2010) the COVR has the ability to discriminate between low and high risk groups. The estimated rate of recidivism was 1.2% for low risk and 63.6% for the high risk group. The observed rates of recidivism in the prospective sample were 9% and 49% for the low and high risk groups respectively.
Contribution to Risk Practice
•The COVR states the risk category of an individual (e.g. ‘low’, ‘high’); however, it does not clarify why an individual is placed at a certain level of risk (Snowden et al., 2009).
•Information taken from patient’s files may be inaccurate or incomplete.
Other Considerations
RATED page updated: July 2019 © Risk Management Authority 2019
that would provide a justification for the statistical information generated by the COVR (Snowden et al., 2009).
•Assessors should note that the COVR has been normed on mentally disordered populations; however, its predictive accuracy lessens depending on the type of recidivism being investigated
•The COVR has been validated for clinical use with acute psychiatric patients who are being considered for release into the community and should be administered by practitioners in mental health disciplines. If the COVR was to be used with other populations, caution should be exercised (Monahan et al , 2005).
•Mixed findings regarding its predictive utility.
•Violent incidents are measured by patient self report, official police records, hospital records, and collateral informants. Total scores are given in a probability format (a percent range for likely violence being committed within the next several months), a frequency format (e.g., for every 100 persons similar to the patient being assessed, between 20 and 32 will commit a violent act over the next several months), and a categorical format (classes of risk, including very low, low, average, high, and very high) (Kennedy et al., 2007).
•Monahan and colleagues (2005) recommend that practitioners adopt a risk assessment procedure that is two fold in nature: administer the COVR instrument; thereafter review the risk estimate generated by the tool. This will allow for additional considerations of risk or protective factors not covered in the COVR assessment. There is also the possibility for conflicting information in the patient’s records. If this cannot be verified, it is advised the practitioner marks the answer as ‘missing’ (Monahan, 2010).
Name of Tool Historical Clinical Risk 20 (HCR 20) developed into HCR 20V3
Category Violence Risk (Validated)
•The HCR 20 prioritises cases as low/routine, moderate/elevated or high/urgent. A low/routine rating suggests the person is not in need or any special interventions or monitoring. Moderate/elevated risk indicates special management and increased monitoring is needed. The high/urgent prioritisation requires immediate action, which could include hospitalisation or suspending a conditional release (Brunt, 2013).
Age Appropriateness
Description
•The HCR 20 is a 20 item structured clinical guide for the assessment of violence risk intended for use with civil psychiatric, community, forensic, and criminal justice populations.
•The third version of HCR 20 (HCR 20V3) was published in 2013 and the encompassing factor on personality now considers all disorder symptoms. The ‘relevance rating’ allows for the rating of the presence and relevance of each risk factor to be evaluated, allowing for assessments to be individualised (Logan, 2014).
Assessors must possess a degree, certificate or licence to practice within health care settings.
•The instrument has a tripartite temporal focus, comprising the following: ten historical variables (‘H’ Scale), looking at a history of problems with violent behaviours and attitudes, employment, relationships, mental and personality disorders and antisocial behaviours; five clinical variables (‘C’ Scale), highlighting recent or current problems with psychosocial, mental health and behavioural functioning; five risk management factors (‘R’ Scale), encompassing relevant past, present, and future considerations with regards to living conditions, services, personal support and stress. All of these scales should be reviewed regularly (Douglas et al., 2014).
18 65 Assessor Qualifications
Strengths
Author / Publisher Webster and colleagues
Assessors must also possess the necessary training and experience in the administration, scoring and interpretation of clinical behavioural assessment instruments and be familiar with professional and research literature in the study of violence. It is possible for a team of professionals to complete the tool: a psychiatrist could complete the items relating to mental illness; a psychologist could look at the personality disorder and psychopathy items; a social worker may complete items pertaining to social history and future plans (Douglas and Reeves, 2010).
Year 2013
RATED page updated: July 2019 © Risk Management Authority 2019
Empirical Grounding
Inter Rater Reliability
•Doyle and Dolan (2006) found ICC values of .85 and .83 for the clinical and risk management items of the HCR 20.
a)UK Research
•Mills et al. (2007) the original HCR 20 achieved an ICC value of .85 in a Canadian sample of incarcerated males.
Validation History
•Gray et al. (2008) ICC of .80 found for the HCR 20V2 .
b)International Research
•Large research base.
•Douglas and Belfrage (2014) found inter rater reliability was evident for the version 3 of HCR 20.
•The HCR 20 has been subject to more than 200 empirical validations (Douglas et al., 2014).
•Cawood (2017) found the inter rater reliability of the HCR 20 V3 was significant with an ICC of .72.
•Douglas et al.’s (2002 2008) review of previous research containing showed ICC value of .73 and above for the HCR 20 across different sample populations.
•The inclusion of a clinical formulation in the HCR 20 exploring the motivating factors for violence and potential future risk scenarios affords the evaluator the opportunity to think about violence in real world scenarios (Brunt, 2013).
•Green et al. (2016): "Results indicated higher inter rater reliability on scoring risk factors among males as compared to females, calling for future research into the role of item indicators across genders and possible differences in interpretations of scoring guidelines."
•Research has shown the HCR 20 includes static and dynamic factors that have sound empirical grounding (Douglas et al., 2005).
RATED page updated: July 2019 © Risk Management Authority 2019
•The HCR 20 has the capacity to guide clinical judgement about intervention and risk management (Gray et al., 2008).
•Doyle et al. (2014) found that "the HCR 20V3 demonstrated very good inter rater reliability and significantly predicted community violence at six and twelve months post discharge, with ROC AUCs of .73 and .70 respectively."
•Coid et al. (2009) the ‘H’ scale generated AUC values of .70 to .73 for female offenders.
a)UK Research
•The HCR 20 was developed from consideration of the empirical literature concerning factors that relate to violence.
•Abbiati and colleagues (2014) applied risk assessment instruments to 52 violent offenders in a Swiss prison to evaluate predictions for physical, any and other misconduct. Total scores were good for physically violent misconduct (AUC=0.80), fair for any misconduct (AUC=0.72) and poor for other misconduct (AUC=0.67).
•Schapp et al. (2009) the HCR 20 score did not predict general and violent recidivism in female psychiatric patients.
Applicability: Females
•Strub, Douglas and Nicholls (2016) study used a sample of 52 men and 48 women "Results indicated that the HCR 20 as well as its components predicted both the occurrence and imminence of violent outcomes and gender did not moderate those relationships."
b)International Research
General Predictive Accuracy
•Garcia Mansilla, Rosenfeld and Cruise (2011) the total score for the ‘H’ and ‘C’ scales had moderate predictive accuracy for community violence (AUC= .60); although when separating the AUC value for the ‘C’ scale alone did not have significant predictive accuracy.
RATED page updated: July 2019 © Risk Management Authority 2019
•The HCR 20V3 was coded alongside other risk assessment tools to check predictive accuracy for 78 female forensic psychiatric patients over a period of 11.8 years. Findings suggest that the HCR 20V3 showed significant predictive accuracy. The clinical scale of the tool was significant for predicting violent recidivism (de Vogel, Bruggeman and Lancel, 2019).
None available at present.
•There are 16 new sub items in the Historical scale in version 3, which prompt the rater to look in more detail at a wider range of historical information (Doyle et al., 2014)
a)UK Research
Validation History
b)International Research
Validation History
Validation History
•Coid et al. (2009) the HCR 20 obtained moderate AUC values for violent recidivism and acquisitive reconviction in male offenders (.67 and .69 respectively). The HCR 20 also generated moderate to high predictive accuracy for female offenders.
•Ho et al. (2009) ROC analyses revealed that the ‘H’ scale had moderate to high predictive accuracy for predicting minor violence (AUC = .619), serious violence (AUC = .74), and any violent incidents (AUC = .61) in a psychiatric sample.
•Lindsay et al. (2008) the HCR 20 obtained a relatively high AUC of .72 in a sample of offenders with learning disabilities.
Applicability: Ethnic Minorities
b)International Research
•O'Shea et al. (2015) maintained that their study demonstrated that "after controlling for a range of potential covariates, the HCR 20 is a significant predictor of inpatient aggression in people with an ID (intellectual disability) and performs as well as for a comparison group of mentally disordered individuals without ID. The potency of HCR 20 subscales and items varied between the ID and comparison groups suggesting important target areas for improved prediction and risk management interventions in those with ID."
•A survey of 43 mental health clinicians in a secure hospital found the historical and clinical subscales of the
•Snowden, Gray and Taylor (2010) the HCR 20 generated moderate to high AUCs of .72 and .66 for White and Black mentally disordered offenders respectively.
RATED page updated: July 2019 © Risk Management Authority 2019
a)UK Research
•Fujii et al. (2005) composite HCR 20 score achieved moderate to high AUC values for native Hawaiian and Euro American groups (.73 and .64 respectively); although for Asian Americans the value was lower (.58). There were no significant differences between AUC values for these ethnic groups.
a)UK Research
Applicability: Mental Disorders
b)International Research
•Jeandarme et al. (2017) carried out a study in 3 forensic medium security units in Belgium. The results indicated that the HCR 20 only shown predictive accuracy for low risk individuals, whilst it was not accurate for high risk patients.
•Campbell, French and Gendreau (2009) meta analysis highlighted the predictive reliability of the HCR 20 in regard to institutional violent recidivism (K = 11, (n = 758) Z+ = .28).
RATED page updated: July 2019 © Risk Management Authority 2019
•Sada and colleagues (2016) utilised the HCR 20 on 225 patients within a Mexican psychiatric facility. It was found that violent behaviour was more severe in the patients within the high risk category, thus suggesting the HCR 20 is a suitable instrument to predict risk of violence.
•The predictive validity of the HCR 20 was examined in a sample of 136 forensic psychiatric patients in Australia. Findings showed that the total score, historical and risk management scales all had moderate to large positive correlations with reconvictions (Shepherd, Campbell and Ogloff, 2018).
Contribution to Risk Practice
•A study by Arai et al. (2016) examined the records of forensic psychiatric patients from 2008 2015 to test the predictive accuracy of the HCR 20. Results from ROC analyses indicate that the clinical and risk subscales of the HCR 20 showed good predictive accuracy, although the historical one failed to do so.
•Vitacco et al. (2016) assessed data from 116 forensic inpatients and found that higher scores in the risk scale of the HCR 20 had a link to a greater likelihood of not being released from or having to return to a forensic facility after release. The authors conclude that clinicians should perhaps consider community based variables when evaluating forensic patients due to be released back into the community.
•Mills et al. (2007) found an AUC value of .73 in their pseudo prospective study of 83 incarcerated males.
HCR 20 were perceived to be the most relevant to violence prediction (Dickens and O’Shea, 2017).
•Dr. Vogel has developed the Female Additional Manual (FAM) which forms an additional supplement to the HCR 20 in relation to assessing violence in women (Vogel et al., 2012; see the ‘Responsivity Section’)
•Few studies have used the categorical risk ratings to determine the predictive utility of the HCR 20 (de Vogel and de Ruiter, 2005).
•The definition of violence provided with the HCR 20 extends to threatened and attempted violence. This means it could be useful to assess risk in cases of violence that do not involve physical harm such as stalking or causing psychological damage (Douglas and Reeves, 2010).
Other Considerations
•The time period for which an assessment is produced needs to be considered. Snowden and colleagues (2007) state that the ‘C’ scale of HCR 20 is found to be a good predictor of institutional violence over the next 3 months but a poor predictor of reconviction over a period of several years.
RATED page updated: July 2019 © Risk Management Authority 2019
•Doyle et al (2014) reports in a study of the third version: "Findings support the hypotheses that (1)the HCR 20 V3 and sub scales can be coded with satisfactory agreement across different raters, and (2) patients with high scores at discharge on HCR 20 V3 were significantly more likely to be violent than service users with low baseline scores at six and 12 months post discharge in the community."
•The focus on mental health and the requirement that the assessor is well versed in mental health interviews is a limitation of the HCR 20 instrument, making it best suited for use with those being managed or moving out of inpatient treatment facilities (Brunt, 2013).
•For more information on HCR 20 (Version 3) please visit: http://kdouglas.wordpress.com/hcr 20/hcr 20/
•The HCR 20 has been translated into sixteen languages and is used across various continents: North and South America, Asia, Europe and Australia (Douglas and Reeves, 2010).
•The HCR 20 should be completed using information obtained from interviews with the individual and other collateral information.
•The HCR 20 can identify a number of risk and responsivity factors relevant to the individual’s risk of violent recidivism.
•Many of the factors identified by the tool can act as targets for treatment/change (e.g. insight, relationship factors) and the instrument can aid decisions regarding the level of monitoring and supervisory strategies, in relation to individuals who pose minimal to high levels of risk for recidivism.
•The HCR 20 does not provide numerical estimates of risk for violence. It is advised that assessors keep abreast of research about the impact of social factors on violence risk and to consider this when applying HCR 20 assessments across various social groups (Douglas and Reeves, 2010).
•The HCR 20 can aid assessors in developing risk formulations and risk management strategies.
•The authors advise that the dynamic items (i.e. the clinical and risk management) are capable of indexing change. In addition, some of the Historical items may not necessarily be ‘static’ (e.g. changes in the offender’s relationship or employment status) (Douglas et al., 2001).
Category Violence Risk (Validated)
•The tool considers strengths rather than being purely risk orientated (Nicholls et al., 2006).
Year 2009
•The tool was initially designed to capture dynamic vulnerabilities and strengths while generating a framework for periodic assessment of risk to inform clinical progress reviews. It should inform treatment, daily management and decision making.
•The tool is intended to assess, document, communicate and manage risk across diverse settings.
Age Appropriateness
Strengths
RATED page updated: July 2019 © Risk Management Authority 2019
Description
•The START includes seven risk estimates which include violence, suicide and self harm. The risk estimates are derived from the consideration of the ratings from the strength and vulnerability scales.
•The 20 items included in the START are drawn from research that have shown these variables to be associated with seven risk estimates/adverse outcomes to individuals with mental health problems and personality dysfunctions, as well as persons who come into conflict with the law.
Name of Tool Short Term Assessment of Risk and Treatability (START)
•Collins et al. (2008) found that clinicians deemed START as appropriate, easy to use and clinically useful.
•The START is intended for use with adults diagnosed with mental, personality and substance related disorders. It is relevant to inpatient and community psychiatric, forensic and correctional populations.
•START is intended for use in both inpatient units and outpatient services.
Assessor Qualifications
•Assessors code the items according to two scales presented in the tool: (1) Strength and (2) Vulnerability.
•The items are rated on a 3 point Likert scale from 0 to 2 and can be coded as both a strength and a vulnerability.
Experienced clinicians from a mental health background. Assessors are required to have participated in relevant training for this tool. It can be completed either by an individual practitioner or jointly by a clinical team via group discussion and reaching a consensus.
•The START is a 20 item structured professional judgement tool designed to structure regular clinical assessments within inpatient and community contexts.
16+
Author / Publisher Webster and colleagues
•Desmarais et al. (2012a) found ICCs of .93 for strength scores, .95 for vulnerability scores and .85 for risk estimates respectively.
•Nicholls et al. (2006) the START attained excellent inter rater reliability (ICC) in various settings; Psychiatry (.80), Nursing (.88) and Social work (.92).
•The START is a concise clinical guide for the dynamic assessment of short term (i.e. weeks to months) risk for violence (to self and others) and treatability.
•Crocker et al. (2011) carried out START assessments on 42 individuals at a civil psychiatric unit in Canada. An inter rater reliability check on six patients six months later found that there was low IRR for total risk score of .38, whereas total strength score was strong at .81.
•Dickens and O'Shea (2015) reported "Inter rater reliability for coding the SOS (Start Outcome Scale) from progress notes was in the excellent range: Cohen's Kappa ranged from .83 to 1.00, the lowest being for self neglect and the highest for self harm and physical aggression."
Inter Rater Reliability
•Timmins, Evans and Tully (2018) assessed the inter rater reliability of START across disciplines, recruiting psychiatrists, mental health nurses, psychologists and occupational therapists to rate 20 case items and 7 risk estimates for two test cases. Good to excellent IRR was found for START items; whilst moderate to poor IRR was found for risk estimates amongst raters. There were clear differences between disciplines at item levels, highlighting the importance of collaborating as a team when completing risk assessments.
•Viljoen et al. (2011) the START strength and vulnerability scale total scores attained good ICC values of .62, and .68.
RATED page updated: July 2019 © Risk Management Authority 2019
Empirical Grounding
b) International Research
•Wilson et al. (2010) found ICCs of .85, .90 and .81 for the strength, vulnerability and risk estimates respectively.
a)UK Research
•The manual claims that the tool is grounded in the HCR 20 and relevant studies of acute violence (Webster et al., 2004). The authors drew upon research from civil psychiatry, forensic psychiatry and corrections reflecting studies from both institutional and community settings.
•Crocker et al. (2011) found that whilst START total risk scores showed good predictive accuracy in relation to physical aggression for periods of 1 and 3 months (AUC ranging from .65 .77), they were not as accurate for the long term of 6 to 12 months. Individuals displaying physical and property aggression had higher risk and lower strength scores on the START.
General Predictive Accuracy
b)International Research
•Braithwaite et al. (2010) suggested there was partial support for the predictive validity of the instrument. Both the strength and vulnerability scales significantly predicted aggression against others and suicidality (AUC= .65 for each scale and behaviour). AUCs of .67 and .63 were generated for substance abuse in the strength and vulnerability scales respectively. Neither scale, however, significantly predicted the occurrence of self harm, suicidality, self neglect or victimisation (AUCs ranging from .52 .58).
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
a)UK Research
•Gray et al. (2011) tested the START in a limited population study of 51 mentally disordered patients. The SPJ scores were able to predict violence to others, verbal aggression, self harm and victimisation (AUCs of .65, .70, ,86 and .67 respectively). The strength and risk scores varied in their ability to predict certain behaviours. The strength scores were poor predictors for all behaviours bar self harm (AUC= .61), with an AUC range of .21 .47. The risk scores were better predictors with an AUC range of .60 .74 for all behaviours; the only exception to this is for self harm which generated an AUC of .48.
•A study by O’Shea, Picchioni and Dickens (2016) of 22 adults in a secure mental hospital found that the inter rater reliability for START items was in the excellent range.
•O’Shea, Picchioni and Dickens (2016) found that the inclusion of strengths improved the predictive accuracy of the START tool. The percentage of cases correctly classified increased from 0.6% to 4.4%. The specific risk estimates scale showed increased predictive accuracy over both the vulnerability and strength scales, showing moderate to large predictive accuracy for all behaviours (AUCs range from .640 .783), bar self neglect (AUC of .546).
Validation History
•Viljoen et al. (2011) in a 3 year follow up in a sample of female forensic patients, the START strength and vulnerability scores showed moderate to large AUCs at .70 and .80 respectively The results show the START scales were predictive of successful reintegration into the community (defined as the absence of readmission to hospital and the presence of an absolute discharge decision) in a sample of female forensic patients.
Applicability: Ethnic Minorities
RATED page updated: July 2019 © Risk Management Authority 2019
•Quinn et al. (2013) found significant predictive validity for adverse incidents at the one month time point and this then diminished over time. Females were rated as having significantly less strengths and more risks than males.
•de Vogel, Bruggeman and Lancel (2019) coded file information for 78 female forensic psychiatric patients using a number of structured professional judgement tools. The START Vulnerability scores showed moderate and large predictive accuracy for all recidivism in medium and long term follow ups (AUCs of 0.748 and 0.698 respectively), as well as for violent recidivism (AUCs of .697 and .704 for medium and long term respectively).
Validation History
•In a study examining aggression data retrieved from institutional records, START strength and vulnerability total scores predicted all forms of aggression, bar physical aggression towards objects, Moderate to large effect sizes were generated for any aggression, verbal aggression and physical behaviours (others) with AUCs ranging from .65 .90. For physical aggression against objects, an AUC of .62 was generated in the strength total score (Cartwright et al., 2018).
b) International Research
a)UK Research
•O'Shea and Dickens (2015) found START was a stronger predictor of aggression and self harm in women than men.
No Empirical Evidence Available
Applicability: Females
•Quinn et al. (2013) discovered that START scores were capable of distinguishing between those with mental disorders at the various stages of their care pathways.
RATED page updated: July 2019 © Risk Management Authority 2019
b)International Research
•Gray et al. (2011) the vulnerability scale was moderately predictive of violence to others (AUC =.68). The strength scale had a significant negative correlation with violence to others (r=. .42) and a corresponding AUC value of .21. Low scores on the strength scale were, thus, predictive of violence.
•Alderman, Major and Brooks (2016) used the START to examine 4559 aggression recordings related to 76 patients with an acquired brain injury. The START risk of violence was classed as low and high for 50% and 13.7% of the sample respectively; suggesting the need for specific tools to be developed for use in patients with ABI.
a)UK Research
•Crocker et al. (2011) carried out a longitudinal study, which indicated that START was well integrated into a Canadian unit's administrative activities.
Validation History
Applicability: Mental Disorders
•Predictive validity of START was evident when Marriott et al. (2017) administered the tool to 527 inpatients within a secure mental health facility in the United Kingdom.
•Wilson et al. (2010) in a 12 month follow up, the strength and vulnerability total scores and the final risk estimates significantly predicted any aggressive acts with AUCs ranging from .82 to .89.
•Chu et al. (2011) in a 1 month follow period, the START vulnerability total scores attained high AUC values in predicting inpatient aggression (.76), interpersonal violence (.78) and verbal threat (.77). Similarly the strength total scores predicted inpatient aggression (.71) and interpersonal violence (.75) but not verbal threat
•Nicholls et al. (2006) START generated moderate to high AUC values for a broad range of aggressive behaviours in a psychiatric hospital: verbal aggression against others (.72), physical aggression against objects
•Braithwaite et al. (2010) the vulnerability scale significantly predicted physical aggression against others (AUC = .66) in a 2 year follow up period.
•Doyle et al. (2008) reported uncertainty over time frame in which risk and strengths are applied from a survey conducted with users of the START.
•The use of the START can help identify factors that are important targets for treatment, intervention and management planning.
RATED page updated: July 2019 © Risk Management Authority 2019
•The START has the ability to create awareness of risk factors and strengths presented by the individual. Findings from previous research also suggest that it may be useful for distinguishing between types of patients (Nicholls et al., 2006; Quinn et al., 2013).
•The START can aid the assessor in examining potential improvement/deterioration in identified risk, responsivity and protective factors which, in turn, can also inform risk management strategies. Further, the tool allows for other harmful scenarios to be considered for individuals, e.g. suicide, substance abuse, self harm and self neglect (O’Shea, Picchioni and Dickens, 2016).
Contribution to Risk Practice
•The START can be completed by a single clinician or by the patient’s multi disciplinary team.
(.67), physical aggression against others (.70) and sexual inappropriateness.
•An electronic START Integrated Treatment Plan (START ITP) has been developed and is being pilot tested in Canada (Leech, personal communication, January 2013).
•Fewer validation studies have been conducted on samples that consist solely of female patients and patients of other ethnic backgrounds.
•Research is ongoing for the START and its use in different settings (e.g. jail diversion programs, Desmarais et al., 2012a).
•O'Shea and Dickens (2015) reported: "The study provides limited support for the START by demonstrating the predictive validity of its specific risk estimates for substance abuse and unauthorised leave. High negative predictive values suggest the tool may be of most utility in screening out low risk individuals from unnecessary restrictive interventions; very low positive predictive values suggest caution before implementing restrictive interventions in those rated at elevated risk."
•Dickens and O'Shea (2015) suggested for lower risk patients assessment at 3 month intervals was appropriate. For those with elevated risk rating more frequent assessments were warranted.
•Repeated assessments using the START can aid assessors in monitoring changes in risk level and identify necessary changes in risk management strategies.
•START is routinely used within forensic mental health populations in the United Kingdom and is recommended by the Department of Health.
•Staff members at a forensic high secure unit in Norway were surveyed about the START. It was felt by 68% of respondents that the existing and potential needs of patients were covered by the tool. Moreover, 73% agreed that using the START tool contributed to a more systematic risk assessment and management process (Kroppan et al., 2011).
•An abbreviated manual is available for use with adolescents (Short Term Assessment of Risk and Treatability: Adolescent Version; START AV) (Nicholls et al. 2010) and the full manual is in preparation by Dr. Viljoen and colleagues. Pilot investigations and other studies have been conducted on the START AV (see Desmarais et al., 2012b; Viljoen et al., 2012).
Other Considerations
•The START includes dynamic factors and the strengths of individuals, which could inform offence analyses and risk formulations.
RATED page updated: July 2019 © Risk Management Authority 2019
•Those using the START tool are to consider any indicators that there are threats of harm that are real, enactable, acute and targeted. Assessors should be mindful of T.H.R.E.A.T in emergency situations where a comprehensive review of the evidence is not possible. (O’Shea, Picchioni and Dickens, 2016).
•Potential limitations of the START tool are it may be too general for certain patients or groups of patients (e.g. those with learning disabilities) (Kroppan et al., 2017).
Description
Category Violence Risk (Validated)
•The Violence Risk Appraisal Guide Revised (VRAG R) is a 12 item actuarial risk assessment instrument for the prediction of violent recidivism among male forensic psychiatric patients.
Strengths
Year 2013
•The instrument was revised in 2013 in order to make it easier to score. Four of the original VRAG items were dropped for using outdated diagnostic criteria or because they have been shown not to be fully applicable to individuals who committed sexual offences (Hertz et al., 2019).
18+
•The VRAG R provides a numerical estimate of the risk of violent recidivism. It is suitable for males aged 18 years and older who have committed serious, violent or sexual offences.
•This tool can be used in combination with historical notes and criminal records (Thomson et al., 2008).
The VRAG was developed from file reviews of 618 male criminal offenders and forensic patients who were initially being assessed for criminal responsibility, fitness to stand trial and/ or being treated in a secure setting; this sample was followed for 7 years. Subsequently, the tool was recalibrated with an extended sample of 800 individuals and followed over a period of 10 years (Quinsey et al., 2006). It was revised in 2013 to make it easier to score. The results of the developmental sample of the VRAG R showed good predictive accuracy with an AUC of .76 (Harris, Rice and Quinsey, 2016).
Age Appropriateness
•The instrument utilises the clinical records as a basis for scoring rather than structured interviews or questionnaires (Harris et al., 2015).
Name of Tool Violence Risk Appraisal Guide Revised (VRAG R)
Author / Publisher Quinsey, Harris, Rice and Cormier
•The VRAG R has a large literature base.
RATED page updated: July 2019 © Risk Management Authority 2019
Assessor Qualifications
Empirical Grounding
Professional expertise and training on instrument.
•Rice, Harris and Lang (2013) developed a revised version of the VRAG (VRAG R), making it easier to score. Both the revised version and the original VRAG yielded high predictive accuracy with an approximate ROC of .75.
b)International Research
•Rossegger et al. (2011) the VRAG obtained an ICC value of .95.
•Doyle and Dolan (2006) found an inter rater reliability value of .99 between three raters based on seven cases.
•Mills et al. (2007) found an ICC value of .95 for the VRAG in a sample of incarcerated Canadian offenders.
RATED page updated: July 2019 © Risk Management Authority 2019
•Olver and Sewall (2018) found the VRAG R displayed excellent inter rater reliability across 35 randomly selected double coded cases, with an ICC value of .97.
b)International Research
Inter Rater Reliability
•Gray et al. (2007) the VRAG obtained a high ICC value of .95
General Predictive Accuracy
a)UK Research
•Using a sample of 296 sex offenders followed up over 17.6 years, Olver and Sewall (2018) found the VRAG R scores demonstrated moderate to large predictive
a)UK Research
None available at present.
•The VRAG has been devised for use in forensic psychiatric settings (see section ‘IV. Mentally Disordered Offenders’). As previously mentioned, other studies have also tested its validity in offenders without psychiatric diagnoses (Langton et al., 2007; Loza and Dhaliwal, 1997).
•Endrass et al. (2008) utilised Krippendorff’s alpha to determine the VRAG’s inter rater reliability. The VRAG attained an excellent inter rater reliability coefficient of .89.
Validation History
•The VRAG was administered to 52 violent offenders in a Swiss prison to test its ability to predict misconduct. The VRAG displayed good predictive validity for physically violent misconduct and any misconduct (AUCs of 0.83 and 0.81 respectively); fair predictive validity was shown for other misconduct (AUC=0.73) (Abbiati et al., 2014).
accuracy for sexual (AUC range=60 .67) and violent (AUC range=.70 .78) recidivism respectively.
Validation History
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
b) International Research None available at present.
•When applied to a sample of 597 male juvenile sexual offenders, the VRAG R showed potential strength in predicting non sexual violent recidivism. It was found, however, that elevated offence severity and adverse childhood experiences encumbered the predictive accuracy of the tool, particularly in the cases of sexual recidivism (Barra et al., 2018).
•In the first European cross validation study of the VRAG R, 534 individuals convicted of a sexual offence were followed up for an average of 7.62 years. The VRAG R showed moderate to large predictive accuracy for violent, general and sexual recidivism (AUCs of .75, .78 and .63 respectively). It was found that predictive accuracy for sexual recidivism was only significant for those convicted of child sexual abuse offences but not for that who committed them against adult victims (Hertz et al., 2019).
•Eisenbarth et al. (2012) the VRAG demonstrated good accuracy in predicting general recidivism in a sample of 80 German female offenders (AUC = .72).
b)International Research
a)UK Research
a)UK Research
Applicability: Females
•Coid et al. (2009) the VRAG generated moderate predictive accuracy of recidivism in a sample of female offenders with ROC values ranging between .65 to .66.
•Snowden, Gray and Taylor (2010) in a two year follow up, the VRAG obtained an AUC value of .74 in predicting violent reconvictions for offenders of Black ethnic origin.
Applicability: Ethnic Minorities
•Hastings et al. (2011) the VRAG was unable to predict institutional misconduct and post release recidivism in female offenders.
b)International Research
Applicability: Mental Disorders
•Kröner et al. (2007) the tool demonstrated moderate accuracy (AUC) in predicting general (.70) and violent recidivism (.70) in a sample of German male offenders undergoing clinical evaluation for criminal responsibility.
•Snowden et al. (2009) the VRAG obtained a ROC value of .77 in a sample of male psychiatric patients.
•Camilleri and Quinsey (2011) report the VRAG has "good predictive accuracy with psychiatric patients of lower intelligence".
•Doyle et al. (2012) the VRAG moderately predicted post discharge violence in a sample of patients discharged from acute mental health units (AUC = .65).
Validation History
•In their retrospective study, Rice et al. (2008) showed that the VRAG had the ability to discriminate risk between non intellectually disabled individuals (control) and intellectually disabled sex offenders.
•Glover and colleagues (2017) tested the VRAG R on a sample of 120 male correctional individuals. Results indicated that the VRAG R gave moderate levels of predictive validity for general and violent recidivism that was able to be sustained over time.
a)UK Research
•A study within an Australian clinical forensic practice found that the revised version of the VRAG (VRAG R) demonstrated predictive validity for recidivism (Brookstein, Daffern and Ogloff, 2016).
•Verbrugge et al. (2011) the VRAG total score attained AUC values of .79 and .92 for violent and general recidivism respectively in a sample of 59 community based offenders with intellectual disabilities.
RATED page updated: July 2019 © Risk Management Authority 2019
•Coid et al. (2009) the VRAG appeared to outperform the HCR 20 and the PCL:R, demonstrating a ROC area of .70 for violent recidivism; the VRAG scores also predicted acquisitive reconviction in males with a ROC of .71.
•Ho et al. (2009) AUC analysis revealed that the VRAG had moderate to high accuracy in predicting minor violence (.70), serious violence (.74) and any violent incidents (.68).
•For more information on the VRAG R, please visit the following website: http://www.vrag r.org/
•Pouls and Jeandarme (2018) collected VRAG scores for 52 offenders with intellectual disabilities (OIDs). AUCs were non significant; although a trend towards significance was evident for physical aggression (AUC=0.74). The results show that the VRAG overestimated risk of OIDs and was only accurate in identifying low risk individuals.
•The VRAG R has the ability to create awareness of static risk factors and can prompt further assessment of the risk of reoffending.
RATED page updated: July 2019 © Risk Management Authority 2019
•Since the VRAG is composed solely of static factors, the tool does not have the capacity to inform treatment protocol or monitor offender progress or motivation for intervention (Daffern, 2007).
•The VRAG R shows some consideration for responsivity issues (e.g. psychopathy).
•In a review of cases decided in United States federal courts, it was found that the VRAG was mainly introduced by the prosecution as a measure of violence risk and was rarely challenged (Cox et al., 2018).
•The VRAG was found to have high concurrent validity with SAQ total scores (Andreau Rodriguez, Peña Fernández and Loza, 2016).
•The tool also relies on PCL:R rating scores as part of the predictive measurement. A study by Doyle, Dolan and McGovern (2002) found that the PCL:SV was a significant contributor to the predictive validity of the VRAG.
Contribution to Risk Practice
Other Considerations
Category Violence Risk (Validated)
Description
•It is two fold in nature consisting of eight proximal warning behaviours (pathway, fixation, identification, novel aggression, energy burst, leakage, directly communicated threat and lastresort behaviour) and ten distal characteristics (personal grievance and moral outrage, framed by an ideology, failure to affiliate with extremist or other group, dependence on the virtual community, thwarting of occupational goals, changing in thinking and emotion, failure of sexual intimate pair bonding, mentaldisorder,creativity and innovation, historyofcriminalviolence)(Meloy et al., 2019).
RATED page updated: April 2021 © Risk Management Authority 2021
•All the proximal warning behaviours are dynamic and based on patterns of behaviour, whilst several of the distal characteristics (e.g. history of mental disorder) are static risk factors. Although protective factors are not explicitly included, the absence of certain indicators (proximal warning behaviours and distal characteristics) are protective. Further, the narrative questions ask about the presence of protective factors in individual cases (Meloy, 2019).
•The TRAP 18 is an SPJ tool for assessing individuals who potentially may engage in lone actor terrorism. It is intended to be used by mental health, intelligence, law enforcement and security professionals to manage operational data on a person of concern and prioritise cases based upon the presenceor absenceofwarning behaviours and characteristics.Itis notan actuarial instrument designed to specifically predict acts of lone actor terrorism (Meloy, 2017).
18+
Age Appropriateness
Author / Publisher Meloy Year 2016
•The focus of the TRAP 18 is on patterns of behaviour rather than distinctive variables, i.e. it is not intended to predict who will or will not commit an act of terrorism; rather, the tool can be used to help assign resources by informing on which individuals should receivepriority attention. Theresults generated from using the tool indicate whether a case requires active management (where one or more warning behaviours are present), or monitoring (where only distal characteristics exist) (Meloy & Genzman, 2016; Meloy, 2018; Meloy, 2019).
•To have the most reliable assessment using the TRAP 18, three sources of data should be used: a direct interview (this may be clinical or non clinical and may or may not involve psychometric testing); collateral interviews with those who are acquainted with the individual and are aware of their behaviour; and theindividual’s public records,includinglawenforcementand nationalsecurity documents if available It is recognised, however, that a direct interview may not be feasible, necessary, or wise in certain cases (Meloy, 2019).
Name of Tool Terrorist Radicalization Assessment Protocol 18 (TRAP 18)
•Meloy et al., (2015) investigated the TRAP 18 using a sample of 22 individuals who had committed terrorism in Europe over a period of thirty five years. Three hundred and ninety six codings were undertaken by two raters who are experts in threat assessment and management. The mean inter rater reliability was found to be0.895. TheIRR range for items was good to excellent ranging from 0.68 to 1.0 for the warning behaviours and 0.75 to 1.0 for the distal characteristics.
•In a study by Challacombe & Lucas (2018), two raters evaluated the whole sample (n=58) using the TRAP 18. Average Cohen’s Kappa was good for proximal characteristics (k=.687) and excellent for distal characteristics (k=.812). The average inter rater reliability for theentire TRAP 18was found tobeexcellent(k=.757).
Assessor Qualifications
•The TRAP 18 is underpinned by theoretical and empirical literature on lone actor terrorism and extremism. Its theoretical underpinnings include theory and research on targeted violence, object relations and attachment theories, gestalt psychology (an attempt to understand meaningful perceptions in a chaotic world system) and psychobiological foundations for predatory violence (Meloy, 2019).
b)International Research
•An overview of the strengths of the TRAP 18 was compiled by experts in risk assessment. It was noted that basing some of the distal factors on psychoanalytic theory provides a clinical understanding that may inform risk assessment and intervention. As a tool, it also has the potential to contribute to the prioritisation of cases in a pre crime scenario, as well as formulation, re formulation and ongoing risk management in a post crime situation (Lloyd, 2019).
Strengths
RATED page updated: April 2021 © Risk Management Authority 2021
Empirical Grounding
Mental health, intelligence, law enforcement, and counter terrorism professionals with caseloads or supervisory responsibilities. Assessors are required to attend the standardised training course lasting 6 7 hours in person or online.
Inter-Rater Reliability
No empirical evidence available at present.
•The tool can be used regardless of ideology (Meloy & Gill, 2016).
a)UK Research
•Another study showed that the TRAP 18 was generalisable across various types of terrorism: jihadists, right wing extremists and single issue attacks (see Meloy & Gill, 2016 for further details).
b)International Research
No empirical evidence at present.
•Erlandsson &Meloy (2018) assessed the2015Swedish school attack in Trollhättan using the TRAP 18. The perpetrator met 7 out of the 8 proximal warning behaviours and 8 out of 10 distal characteristics on the TRAP 18 instrument. Based on these results, the authors
•After examining 111 lone actor terrorist attacks in the United States and Europe, Meloy & Gill (2016) found that the TRAP 18 was able to discriminate between lone actor terrorists who successfully carried out their attacks compared to those whose attacks were thwarted. The five variables found to be significantly different were fixation, creativity/innovation, failure in sexually intimate pair bonding, pathway and less likely to be dependent upon a virtual community.
•Böckler et al., (2015) used the TRAP 18 to assess the case of the 2011 Frankfurt Airport Attack. Carrying out a qualitative analysis of investigation and court files found that the perpetrator showed six proximal warning behaviours and nine distal characteristics. Tracing the various stages of the individual’s life highlighted several triggers towards him drawing upon jihadist ideologies.
RATED page updated: April 2021 © Risk Management Authority 2021
a)UK Research
General Predictive Accuracy
•There have been a number of validation studies carried out on the TRAP 18. An examination of the postdictive validity of warning behaviours was first carried out by looking at school shooting case studies. Although not a form of terrorism because school shootings are usually not motivated by political or ideological reasons, these attacks are similarly unpredictable and have the potential to cause mass causalities (Meloy et al., 2014).
Validation History
•Challacombe & Lucas (2018) applied the TRAP 18 to a series of violent and non violent incidents involving Sovereign Citizens in the US. Chi square and binary logistic regression analyses wereused to test theability of the TRAP 18 to predict violent outcomes. The full model was statistically significant (x2=33.88), suggesting TRAP 18 was able to distinguish between individual cases that were violent and non violent.
Validation History
a)UK Research
b) International Research
b)International Research
No empirical evidence at present.
Applicability: Mental Disorders
Applicability: Ethnic Minorities
No empirical evidence at present.
concluded there is an excellent goodness of fit between this incident and other cases of individual terrorism in Europe and North America.
RATED page updated: April 2021 © Risk Management Authority 2021
Applicability: Females
No empirical evidence at present.
a)UK Research
b)International Research
No empirical evidence at present.
•Goodwill & Meloy (2019) used a combined sample of North American terrorist attackers (n=33) and non attackers (n=23) to plot the potential clustering (co occurrence) of risk factors. Findings indicated that proximal warning behaviours are present in attackers and largely absent in non attackers, whilst distal characteristics are evident in both groups. Three of the distal characteristics (personal grievance and moral outrage, ideological framing, and changes in thinking and emotion) cluster with both the proximal warning behaviours and the attackers. This suggests both that these distal factors co occur more in attackers than non attackers, and that there is an increased likelihood of finding proximal warning behaviours than any of the remaining seven distal characteristics
No empirical evidence at present.
a)UK Research
•Fernández García Andrade et al. (2019) applied the TRAP 18 to 44 patients with severe mental illness, who had a criminal history and were in situations of social exclusion. High predictive validity was demonstrated for the TRAP 18 (AUC=1.00), indicating it could be a useful tool for assessing the risk of terrorist radicalisation in
Validation History
Validation History
•Meloy et al., (2019) coded 2 non random samples of convenience: 33 cases of a lethal terrorist attack in the United States; 23 individuals who posed a national security concern but did not mount an attack: the latter group were either successfully risk managed for three years, or had no intent to mount an attack. Half of the TRAP 18 indicators were found to be significantly different between the samples with medium to large effect sizes (ϕ=.35 .70). The three warning behaviours that were not significantly different between the groups were fixation, novel aggression and leakage. Due to its retrospective design, noinferencesweremadeaboutpredictivevalidity on thebasis ofthis study.
Contribution to Risk Practice
•In a review of the literature, Guldimann & Meloy (2020) described the inter rater reliability of the TRAP 18 as excellent, with research showing promise in terms of content, criterion, discriminative, and predictive validity. They found that several of the proximal warning behaviours pathway, fixation, identification, leakage, energy burst, and last resort were commonly found in the research, while “directly communicated threat” was not prominent. However, they caution that its absence should not be interpreted to mean that no threat exists.
•A group of experts reviewed the strengths and limitations of the TRAP 18 Some limitations noted were that:
•In terms of the strengths of the tool, they noted that:
Other Considerations
•Brugh et al., (2020) applied the TRAP 18 to a sample (n=77) of jihadism inspired lone actor terrorists in Europe and the US, using only publically available information from the Western Jihadism Project database. Of 18 items, only four were rated Present more often than they were rated Absent or Unknown (Pathway, Identification, Personal Grievance, Framed by Ideology). In comparing the US and European samples, the items Fixation, Energy Burst, Leakage, and
o some of the more psychoanalytic distal factors may be difficult to make sense of without clinical expertise.
o it can potentially assist with case prioritisation in a pre crime scenario, in addition to formulation, re formulation and ongoing risk management in a post crime situation; and
o a clinical understanding of subjects relevant to terrorism underpin and inform the tool (Lloyd, 2019).
•Tools like the HCR 20 V3 and WAVR 21 could function as a ‘gateway’ instrument to allow for a more individualised assessment of the behaviours and motivations associated with lone actor terrorism using the TRAP 18 (Meloy, 2018). Guldimann & Meloy (2020) suggest using other tools such as the HCR 20 V3 in conjunction with the TRAP 18.
o the focus on lone actors potentially limits its utility;
•Looking at a sample of 22 individuals who had committed terrorism in Europe, Meloy et al. (2016) found that ‘content validity’ (the extent to which a measure represents all the facets of a given construct) was evident in 72% of the variables of the TRAP 18.
RATED page updated: April 2021 © Risk Management Authority 2021
mentally ill individuals, particularly those with a history of being in prison and living in socially secluded situations.
o a full assessment involving a direct interview, psychometric testing, and complete information sources may not be entirely realistic in a pre crime scenario; and
o it can be used in risk management and prevention, potentially discriminating between empty and real threats;
•Training in the TRAP 18 is available from the Global Institute of Forensic Research through their online on demand resources (gifrinc.com). This company is owned by Multihealth Systems.
•Further information about the tool and its author may be found here: http://drreidmeloy.com/training/trap 18/
RATED page updated: April 2021 © Risk Management Authority 2021
Dependence on the Virtual Community were more common the US sample. The study produced three “false negatives:” three cases were not recommended for Active Risk Management, contrary to expectations for a sample of confirmed lone actor terrorists. The authors conclude that using TRAP 18with only publically available information raises questions aboutthetool’s feasibility in this setting. Consideration should also be given to potential difficulties using the TRAP 18 across different geopolitical contexts where there may be differences in how information is gathered and made publicly available.
•The TRAP 18 is owned, copyrighted and trademarked in the United States by Dr Meloy, with distribution and sales licensed to Multihealth Systems, Inc. (mhs.com) (Meloy, 2019).
•The DRAOR contains dynamic stable, acute and protective factors that assist staff in identifying and responding to change in the risk of reoffending and desistance from crime.
Name of Tool Dynamic Risk Assessment of Offender Re Entry (DRAOR)
•The DRAOR is scored using information obtained from interviews with the individuals, their families and partners, reports from treatment providers and other applicable persons such as the police (Yesberg, Scalan and Polaschek, 2014).
Age Appropriateness
Assessor Qualifications
Year 2009
RATED page updated: July 2019 © Risk Management Authority 2019
Author / Publisher Serin and Mailloux
•The DRAOR was firstly developed for probation staff in Canada to be able to manage the management of those in the community. Serin and colleagues extrapolated items from previous research on violent and sexual offending (Serin, Lloyd and Hanby, 2010).
•Polaschek and Yesberg (2018) looked at completers of an intensive prison based treatment programme. Based on DRAOR scores, completers entered the community with higher protective and lower stable and acute dynamic factors and also showed less variability on acute risk factors.
•R. Serin (personal communication, October 2010) initial validation research in New Zealand involved 283 individuals who offended and 35 probation officers. The researchers found that the DRAOR attained moderate to high accuracy in predicting recidivism: ‘Stable’ subscale scores in predicting any new offence (AUC = .75), ‘Acute’ subscale (AUC= .76). It was also found that repeating the DRAOR assessments resulted in better prediction. The first set of Stable scores attained an AUC value of .70 which increased to .79 by the fourth iteration. The tool was further refined and thereafter adopted as a national standard for probation in New Zealand (Serin, Lloyd and Hanby, 2010).
•Tamatea and Wilson (2009) found positive correlations between the Stable and Acute subscales and the risk of recidivism. The Protective subscale was negatively correlated to recidivism.
•The basis for this instrument is the ‘risk needs responsivity’ model, which categorises interventions based on risk and changing needs (Yesberg et al., 2015).
No age range specified
Description
Category Violence Risk (Awaiting Validation)
The total score for the DRAOR is calculated by adding the stable and acute scores and then subtracting the protective total (Yesberg and Polaschek, 2014).
Designed for use within the probation service. No assessor qualifications specified.
Tool Development
•A Masters dissertation administered the DRAOR to 85 individuals convicted of sexual offences released from prison to test its ability to predict sexual, violent and general recidivism. It was found that the domain scores significantly correlated with all recidivism, bar sexual offending (Averill, 2016).
•A doctoral thesis by Hanby (2013) found that the DRAOR has promise as a valid tool for risk assessment and management, with reconvictions being accurately predicted from monthly average Stable Risk for 12 months and the Protective Factors were predictive for 4 months.
•A sample of 287 high risk males discovered that the DRAOR total score significantly predicted reconviction. Further, the score of stable items was found to independently predict reconvictions and imprisonment, indicating that this is the most important element to the tool in predicting outcomes (Yesberg and Poalschek, 2014).
• As part of Masters dissertation research, Chadwick (2014) sought to validate the DRAOR. Predictive accuracy was evaluated using ROC analysis. Findings indicated the stable domain and total scores produced the largest effects (AUCs of .61 to .62). It is concluded that case managers would benefit from utilising the DRAOR in the everyday supervision of individuals who have offended.
•Further validation is being undertaken in Australia, Canada and the United States (Serin, Lloyd and Hanby, 2010).
•Positive reports of the application of DRAOR from Department of Corrections New Zealand 2013.
•Serin (personal communication, December 2012) a 3 month follow up pilot study conducted in Iowa, utilised a total of 926 DRAOR assessments obtained from individuals within probation and parole settings. Data was extracted at two time intervals, with a mean of 65.6 days between assessments. Small to moderate correlations was observed between the composite and subscale scores at time 1 and the outcome which was defined as any violation of return to prison (Total score = .28; Acute score = .23; Stable .25; Protective score = .28).Similar trends in correlation were observed at time 2. Furthermore, the DRAOR attained moderate AUC values in predicting recidivism (Total score = .66, Stable = .60, Acute = .65, Protective = .67) compared to the LSI R (.56).
•Authors advise that the assessment should be completed on a monthly basis.
RATED page updated: July 2019 © Risk Management Authority 2019
•Research by Yesberg and colleagues (2015) on a mixed sex sample found that the DRAOR predicted recidivism for the females but not the males.
General Notes
•Using a sample 112 males convicted of IPV offences, a Masters dissertation looked at whether the DRAOR predicted repeat offending. While the DRAOR did not predict IPV recidivism in this sample, it appeared to be useful for informing case management decisions. IN terms of violence generally, the DRAOR’s Total and Stable scores were significantly positive associated with offending (AUC=.65); whilst the Protective scores were negatively associated with violence (AUC=.35) (Perley Robertson, 2019).
RATED page updated: June 2019 © Risk Management Authority 2019
•In addition to making a final judgment on the risk of violent behaviour towards others, the FAM also allows for the individual to be assessed for their risk of self destructive behaviour, victimisation and non violent criminal offending (de Vogel, Wijkman and de Vries Robbé, 2018).
Category Violence Risk (Awaiting Validation)
Name of Tool Female Additional Manual (FAM)
Age Appropriateness
•The FAM is a tool with additional guidelines to the HCR 20V3 that aids clinical assessment of violence risk in adult females who have committed prior violence offences. Similar to the HCR 20 / HCR 20V3, the FAM covers historical, clinical and risk management items (de Vogel et al., 2014).
Author / Publisher de Vogel, de Vries Robbé, van Kalmthout and Place
Year 2014 (2012)
Tool Development
The historical items are personality disorder, traumatic experiences, prostitution, parenting difficulties, pregnancy at a young age and suicidality/self harm. The clinical items are covert/manipulative behaviour and low self esteem. The risk management items consist of problematic childcare responsibility and problematic intimate relationship.
•The FAM items were constructed following a thorough review of the literature and clinical experience in relation to females. The nine new risk items reflect gender responsive issues such as problems with childcare responsibilities, prostitution, low self esteem and covert / manipulative behaviours (de Vogel, Wijkman and de Vries Robbé, 2018).
Assessor Qualifications
Assessors must also possess the necessary training and experience in the administration, scoring and interpretation of clinical behavioural assessment instruments (de Vogel et al., 2014)
•The FAM was developed due to there being a lack of gender specific tools to be used for violence in female populations. A literature review, interviews with mental health professionals and a pilot study in a Dutch mixed gender forensic psychiatric hospital were used to develop the FAM (de Vogel, Wijkman and de Vries Robbé, 2018).
•The FAM is comprised of additional guidelines to five of the historical HCR 20V3 items or two HCR 20V3 itemsand eight additional risk items specifically for evaluating females who have offended (de Vogel et al., 2014)
18+
Assessors must possess a degree, certificate or licence to practice within health care settings (de Vogel et al. 2014)
Description
•The FAM assessor is invited to decide upon risks for various behaviours and scenarios: risk for future violence (influencing someone else to commit violence or being an accessory to violence is also included in the definition of violence); risk for serious physical harm; risk for imminent violence; risk for self destructive behaviour; risk for victimisation; risk for non violent criminal behaviour (de Vogel et al., 2014)
•In the manual, the developers maintain that the FAM may be possibly be partly useful for violence risk assessment in adolescent girls. This is said with caution, however, because there are some risk factors specifically valid for adolescent girls that are not included in the FAM, such as interaction with deviant peers, being a member of a gang and running away from home. Further to this, some of the FAM items are not applicable to adolescent girls, such as ‘victimization after childhood (de Vogel et al., 2014). It is, therefore, recommended that scholars consider developing or adapting tools for risk in adolescent girls, as well as for assessing the risk of child abuse, intimate partner violence or psychopathy in female populations (de Vogel, Wijkman and de Vries Robbé, 2018).
•de Vogel, Wijkman and de Vries Robbé, 2018 suggested the FAM could be useful in general mental health settings and for detecting inpatient violence.
RATED page updated: June 2019 © Risk Management Authority 2019
•In 2011, the tool was revised subsequent to user feedback, new insights and experiences with coding of other tools, specifically the Structured Assessment of Protective Factors for violence risk (SAPROF) and the Short Term Assessment of Risk and Treatability (START). In addition to a final judgement on risk to others it now also includes judgements on self destructive behaviour, victimization and non violent criminal behaviour. There was a further revision in 2013 so it could be used with the HCR 20V3 (de Vogel, Wijkman and Vries Robbé, 2018).
•A draft version of the FAM was implemented in 2007 for all female patients who were resided in the Van der Hoeven Kliniek, a Dutch forensic psychiatric hospital admitting both men and women. Interviews with mental health professionals revealed there were factors particularly relevant to women: covert behaviour, hiding or concealing the truth, manipulative use of sexuality such as exploiting it for personal gain and low self esteem (de Vogel and de Vries Robbé, 2013).
•Griswold and colleagues (2016) used a sample of 28 female defendants adjudicated not guilty by reason by insanity in the United States. The FAM showed good inter rater reliability and predictive validity for inpatient violence. In spite of this, the authors found no incremental validity of the FAM over the HCR 20.
•A study by Greig (2014) into the psychometric properties of the FAM led to the conclusion that the tool may be useful in civil psychiatric populations.
•In 2011, research was conducted on the psychometric properties of the FAM in the Van der Hoeven Kliniek. The authors found good inter rater reliability for the composite score and the different final risk judgements, except for victimization (de Vogel et al., 2011). Preliminary findings on the FAM’s predictive validity during treatment were good for incidents of violence towards others as well as for incidents of self destructive behaviour.
•Campbell and Beech (2018) examined whether scores on the HCR 20 and FAM can be related to frequency of self harm in 89 female psychiatric patients. The association between self harm and HCR 20 scores was strengthened by the inclusion of the FAM. It is recommended that the FAM is used alongside the HCR 20 when assessing risk of self harm in females.
•de Vogel et al. (2019) coded file information in 78 female forensic patients using a number of risk assessment instruments, including the FAM. Reconviction data was available for 71 of the patients. The FAM was one of the tools showing the highest predictive accuracy for all recidivism (including violence).
General Notes
RATED page updated: June 2019 © Risk Management Authority 2019
•Enquiries regarding this risk assessment tool can be sent to the following e mail address: vdevogel@dfzs.nl
•Although it is preferable to use the most recent edition of the FAM (de Vogel et al., 2014), the original FAM can also be applied as an additional manual to the HCR 20V3 with some adaptions.
•Considering risk levels and protective factors may fluctuate over time The developers of SAPROF recommend that repeated assessments are carried out if there are changes in an individual’s circumstances (de Vogel et al., 2011).
Name of Tool Structured Assessment of Positive Factors for Violence Risk (SAPROF)
Age Appropriateness
No age range specified.
Author / Publisher de Vogel and colleagues
Experience and training in the administration and interpretation of tests and semi structured Assessorsinterviews.should also be familiar with the most recent professional and research literature on the causes and prediction of violence.
Tool Development
•The SAPROF is a 17 item checklist based on the structured professional judgement methodology. It was developed to be used in a variety of settings: forensic and general psychiatry (both inpatient and outpatient); prisons and probation supervision.
The most important items can be highlighted by marking factors as ‘keys’ most likely to offer protection or ‘goals’ that should focus on improvement for individuals.
•The SAPROF was designed to complement other SPJ risk assessment tools such as the HCR 20 or the HCRV3 by considering both protective and risk factors in evaluating risk assessment for future violence.
•The tool considers the importance of dynamic factors and their role in achieving effective treatment programmes (de Vogel et al., 2009).
Assessor Qualifications
Description
•The items are categorised into 3 subscales: (1) internal items, characteristics that could offer protection against future violence; (2) motivational items, which encourage the individual to be a positive member of society; (3) external items, those environmental factors that could be of benefit.
•The English version of the SAPROF was made available in April 2009; the second edition of the tool was published in 2012.
Category Violence Risk (Awaiting Validation)
Experience and training in conducting individual assessments.
RATED page updated: July 2019 © Risk Management Authority 2019
Year 2009, 2012
• The SAPROF was initially developed for the forensic psychiatric population in 2007.
General Notes
•The SAPROF is to be used alongside SPJs like the HCR 20 and may also be used with actuarial tools. It is highlighted that it is imperative to consider the circumstances of every individual, for every item on the SAPROF may also be a risk factor as well as a protective one (de Vogel et al., 2011).
•In a study of 52 individuals convicted of violent offences in a Swiss prison, SAPROF total scores showed good predictive validity for physically violent misconduct with an AUC of 0.84. Poor predictive validity was found for any misconduct and other misconduct, yielding AUCs of 0.64 and 0.61 respectively (Abbiati et al., 2014).
•Yoon, Spehr and Briken (2011) in a pilot study conducted with a German sample of individuals with sexual offences, the SAPROF had a significant negative correlation with other risk assessments such as the STATIC 99.
•The authors maintain that whilst clinical use of the SAPROF is possible, results should be interpreted with caution. An update to caution statement was provided in January 2016: “Given the strong empirical findings regarding the psychometric properties of the SAPROF, in particular its inter rater reliability, predictive validity for desistance in those with violent as well as those with sexual offending histories and the demonstrated relation between improvements in protective
•de Vries Robbé, de Vogel and Douglas (2013) coded the HCR 20 alongside the SAPROF on a sample of 188 patients who had been discharged from forensic psychiatric treatment. It was found that combining the risk factors of the HCR 20 with the protective factors of the SAPROF resulted in good predictive validity for violent recidivism after treatment.
•A comparison of various risk assessment tools found that the SAPROF has good ‘construct validity’ with the START tool, showing that both instruments measure the same thing (Abidin et al., 2013).
•de Vries Robbé, de Vogel and de Spa (2011) SAPROF obtained an excellent ICC value of .85 for the composite score. SAPROF demonstrated high predictive accuracy in relation to post discharge recidivism in a sample of forensic patients (AUCs .74 .85).
•de Vries Robbé and de Vogel (2012) the authors cite the findings of studies in preparation for publication, which investigated the predictive accuracy of the SAPROF with those convicted of both violent and sexual offences. The investigations found high AUC values ranging from .71 to .85 in three follow up periods (1 , 3 and 11 year follow ups) for the composite score. These trends were observed across reconvictions for violent and sexual offences.
•Yoon et al. (2016) retrospectively rated 450 individuals convicted of sexual offences in Austria. The inter rater reliability was shown to range from good to excellent for all SAPROF items; whilst the predictive accuracy was found to be low to moderate for various types of recidivism.
RATED page updated: July 2019 © Risk Management Authority 2019
•A mixed sex sample of 409 patients discharged from medium secure services in England and Wales utilised both the HCR 20 V3 and the SAPROF at six monthly intervals to test their predictive accuracy in determining which patients would carry out violence within the first year of release. It was found that only a few items in the SAPROF demonstrated any discriminative value in identifying which patients would not engage in violent behaviour (Coid et al., 2015).
•Research carried out on 83 individuals convicted of sexual offences found that the SAPROF had good predictive validity for sexual violence (de Vries Robbé et al., 2015).
•Kashiwagi et al. (2018) examined the inter rater reliability and predictive accuracy of the SAPROF in 96 patients located in forensic mental health units in Japan. Since there are no widely used structured risk assessment tools for violence in Japan, the SAPROF was translated into Japanese. Moderate to good inter rater reliability was evident, with an ICC of 0.70 for 30 randomly selected cases. The predictive accuracy was an AUC of 0.87 and 0.85 for 6 and 12 months respectively.
•Enquiries regarding this risk assessment tool can be sent to the following e mail address: saprof@hoevenkliniek.nl
•Fifteen translations of the tool are now available. Other validation research is currently underway in Canada, New Zealand, Germany, Italy, Switzerland, Portugal, Ireland and the UK (de Vries Robbé and de Vogel 2012)
•Research and training enquiries can be made by contacting the authors using the following email address: mdevriesrobbe@hoevenkliniek.nl
•For more information about the tool please visit: www.saprof.com
The SAPROF Intensive Care highlights additional factors that are particularly relevant in an inpatient or forensic psychiatric care setting.
A SAPROF Youth Version (SAPROF YV) has been developed for use with youths with violence related problems. The intention is for the SAPROF YV to be used in conjunction with risk focused youth tools such as the YLS/CMI or the SAVRY. This was implemented nationally across juvenile justice institutions in the Netherlands; it is be used in addition to the SAVRY tool (de Vries Robbé and Willis, 2017). The SAPROF Sexual Offending (SAPROF SO) is to be used to assess protective factors specific to sexual offending.
factors and recidivism reduction, this tool may be used as a risk assessment and treatment guidance tool in clinical practice as well as research.”
•A number of other variations of the SAPROF are currently being developed (interested readers are directed to de Vries Robbé and Willis, 2017):
•The authors have published an online article detailing the differences between the first and second versions of the SAPROF manual. This includes (but is not limited to) further notes regarding the time frames for the coding of SAPROF items, updates in SAPROF validation research and its application to practice and risk formulations.
RATED page updated: July 2019 © Risk Management Authority 2019
For those with intellectual deficits, the SAPROF Intellectual Disabilities (SAPROF ID) is in progress of being designed.
•The primary purpose of the WAVR 21 is to examine the risk for homicide; a secondary element to the tool is to assess the risk, frequency and severity of other workplace aggression such as stalking (Meloy, White and Hart, 2013).
Name of Tool Workplace Assessment of Violence Risk (WAVR 21 V3)
RATED page updated: July 2019 Authority 2019
Year 2016
•The WAVR 21 was first published in 2007; it is now in its third version which came out in 2016 (see WAVR 21 webpage)
•Printed manuals are available at specializedtraining.com; training opportunities are regularly posted at wtsglobal.com. One and two day trainings are available.
•The intended users of the tool are qualified mental health professionals with experience of assessing violence risk and knowledgeable about workplace legal issues, or members of multidisciplinary threat assessment teams (Meloy, White and Hart, 2013).
•The WAVR 21 breaks down an individual’s thinking about violence into three categories: motives for violence; homicidal ideas, violent fantasies or preoccupation; violent intentions and expressed threats (Brunt, 2013).
© Risk Management
Author / Publisher White and Meloy
•Using the WAVR 21 requires training, a knowledge of threat assessment literature and risk management principles, and compliance with relevant local and national laws (Brunt, 2013).
Age Appropriateness
18+ Assessor Qualifications
•The existing literature on workplace violence, threat assessment and risk was reviewed to formulate the WAVR 21. It was pertinent that the tool captured both escalation to targeted or intended violence, as well as de escalation where an individual ultimately decides against the violence they had previously contemplated (Meloy, White and Hart, 2013).
Category Violence Risk (Awaiting Validation)
Tool Development
Description
•The WAVR 21 is the first and only SPJ instrument designed to investigate the risk of workplace and campus related targeted violence, i.e., lethal situations where an individual enacts violence against a specific target in order to cause as much harm as possible (Brunt, 2013; Kienlen, n.d.).
•The WAVR 21 is a 21 item instrument examining violent motives, ideation, intent, weapons skills, pre attack planning, negative personality traits, mental disorders, situational factors and any protective factors. Items are coded as absent, present, or prominent, with an additional “recent change” determination (Brunt, 2013; Kienlen, n.d.; Meloy, White and Hart, 2013).
•Participants who take part in formal WAVR 21 training are provided with an additional tool, the ‘PROTECT’ form, used to identify stabilisers against violence risk (Kienlen, n.d.).
General Notes
•The focused questions at the end of the WAVR 21 coding sheet allow for further exploration of issues pertinent to threat assessment (Brunt, 2013). The intake documentation form in the WAVR 21 V3 provides for an initial assessment for prioritising of the case.
•Further details about the tool can be found here: https://www.wavr21.com/
•Scalora, Cawood and Viñas Racionero (in press) tested the predictive validity of the WAVR 21 using forty cases of violence that had taken place in workplaces and academic institutions. Raters were blind to the known outcomes. Substantial predictive validity was demonstrated with an AUC of .70, showing that final summary risk ratings correlated with physical violence, and correct classification of cases as either violent or nonviolent was comparable to other structured professional judgment instruments.
•Typical users of the WAVR 21 are members of multi disciplinary threat assessment and management teams or mental health professionals who consult or conduct formal assessments in work or campus settings. Other potential users of the WAVR 21 are workplace violence security consultants, as well as law enforcement professionals who assist the organisations in their communities (see WAVR 21 webpage).
•A digital version of the WAVR 22 V3 is available in two forms: an online tool as part of the Resolver platform (resolver.com); and licensing as intellectual property by White and Meloy so the WAVR 21 V3 can be formatted according to the wishes of the licensee and used behind a secure firewall.
RATED page updated: July 2019 © Risk Management Authority 2019
•Eleven raters assessed 12 cases of workplace threat situations chosen at random. The inter rater reliability was found to be excellent for two items, fair to good for eleven and poor for eight of them. This may be countered, to some extent, by the fact that the raters having limited experience and no mental health background; the case materials were limited in their quality and quantity. The overall sum of the risk factors generated a good IRR of .67 (Meloy, White and Hart, 2013). When psychologists alone utilised the WAVR 21, interrater reliability was in the excellent range.
specification
Extremism Risk Guidelines (ERG22+)
Name of Tool
The tool does not include a list of protective factors, but assessors are encouraged to consider whether each factor (or its absence) can act protectively.
Assessors rate items as Present, Partially Present, or Not Present, then develop a formulation to explore the relevanceand function ofeach indicator nowand in thefuture, both in thecontext of the three domains as well as in disengagement or desistence. (Powis, Randhawa Horne & Bishopp, 2019; Lloyd & Dean, 2015; Logan & Lloyd 2018).
The ERG22+ process includes an interview or written comments from the individual being assessed. However, it is possible to complete an ERG22+ without this (HMPPS, 2019)
Description
The tool includes 22 factors categorised into three domains: Engagement, Intent, and Capability
Assessors should use as many sources of information as possible, including interviewing family members in some cases, and work collaboratively with the individual to understand their pathway into extremist offending (HMPPS, 2019).
No age
The “+” in ERG22+ represents the ability to add factors that are relevant for a given individual.
Identifying more risk factors as “Present” does not in itself indicate a higher risk; rather, assessors need to explore how these factors combine to tell an individual’s “risk story” (Lloyd & Dean, 2015, p.13).
The tool is not limited to a specific extremist ideology, and has been used with individuals associated with Islamist, animal rights, far right, far left, and gang affiliated groups (Dean et al., 2018; HMPPS, 2019).
Age Appropriateness
RATED page updated: April 2021 © Risk Management Authority 2021
The tool has been used in risk assessment and management, including sentence planning and decisions relating to parole, relocation, recall, and licence conditions (HMPPS, 2019; Lloyd & Dean, 2015).
Year 2011
.
The tool has been integrated into several interventions in England and Wales, such as the Healthy Identity Intervention (HII) and the Motivational and Engagement Intervention, which were informed by the tool’s 22 risk factors. All participants have an ERG22+ assessment completed, the outcome of which guides the interventions used (Dean, 2014; Herzog Evans, 2018; Dean et al., 2018). These interventions have since been amalgamated into Healthy Identity Intervention: Foundation and Healthy Identity Intervention: Plus (Dean et al., 2018).
The Extremism Risk Guidelines (ERG) 22+ is a structured professional judgement (SPJ) tool for assessing the risks and needs of those convicted of terrorist extremism offences, which may or may not include violent extremism (HMPPS, 2019).
Category Violence Risk (Awaiting Validation)
The tool does not generate a score or a risk banding system (van der Heide et al., 2019).
The individual items are dynamic, except for criminal history (Lloyd & Dean, 2015).
Author / Publisher Her Majesty’s Prison and Probation Service (HMPPS)
Assessors must be chartered and registered psychologists, or experienced probation officers. They must work in a role that requires assessment of convicted extremist offenders and/or those for whom there is credible concern about their risk of extremist offending. (HMPPS, 2019).
Currently, use of ERG22+ is only licenced within HMPPS.
The tool was developed by HMPSS (formerly National Offender Management Service [NOMS]).
Powis, Randhawa Horne & Bishopp (2019) applied the tool to a sample (n=171) of convicted Islamist extremists and used multidimensional scaling analysis (MDS) to examine construct validity and internal consistency reliability. The tool had good internal consistency overall, with an alpha coefficientof 0.80. Individualdomains had varying results:theEngagementand Intent domains showed moderate internal consistency (alpha coefficients of 0.65 and 0.79, respectively). The Capability domain showed low internal consistency (0.46). The authors concluded that the 22 factors may be better organised into different domains (they suggest Identity & External Influence; Motivation & Ideology; Criminality; Capability; and Status and Personal Influence as preliminary categories) rather than the existing domains of Engagement, Intent, and Capability. They also suggest the mental health item could be refined and defined further.
Assessors must undertake a 2 day training course. (HMPPS, 2019).
It is desirable if assessors also have experience with psychologically informed risk assessment and formulation. (HMPPS, 2019).
General Notes
RATED page updated: April 2021 © Risk Management Authority 2021
Herzog Evans, (2018) concluded that the ERG22+ may work best in countries with a legal threshold for terrorism offences that includes non violent terrorist acts, such as in the UK and France.
The ERG22+ was informed by casework (approximately 30% of the convicted terrorist population at the time), the body of literature, and research commissioned by the UK government (Lloyd & Dean, 2015)
Tool Development
Powis, Randhawa Horne & Elliot (2019) tested the inter rater reliability of the tool using two formats: to test research reliability, two experienced raters assessed 50 randomly selected
The ERG22+ was developed from an earlier Structured Risk Guidance (SRG) protocol (2009). The SRG was developed from casework with convicted extremist offenders and the literature on terrorism. Following a pilot phase and independent evaluation, as well as feedback from assessors, offenders, stakeholders, and peer reviewers, and developments in the body of literature, the SRG was revised and developed into the ERG22+ in 2011 (Dean et al., 2018)
An ERG22+ assessment is carried out for every individual convicted under terrorism legislation in England and Wales (Powis, Randhawa Horne & Bishopp, 2019).
The ERG22+ was developed to be applicable to extremist offenders with or without a history of violence. The developers note that the majority of those with terrorist convictions in the UK do not have a history of violent convictions (Lloyd & Dean, 2015).
Assessor Qualifications
The tool is adaptable to any individual, regardless of age or gender. The factors are psychological, requiring a qualified assessor to interpret and apply them to a given individual. However, the developer notes that the tool may not be suitable in certain contexts, such as jihadi children taken abroad by their parents who are returning to the UK (M Lloyd, personal communication, 22 February 2021).
A group of experts summarised the strengths and limitations of the tool. Strengths include:
o that it is completed collaboratively with the individual;
cases; to test field reliability, 33 raters of varying experience assessed two specially developed test cases, which were then compared to “gold standard” ratings. The inter rater reliability of the research reliability test was in the ‘excellent’ range, while the field reliability test had more varying results: the Intent domain had poor inter rater reliability, while the engagement and capability domains were ‘moderate’ to ‘borderline good.’ Experienced raters had higher levels of agreement. The authors conclude that additional training and clearer definition of terms could improve inter rater reliability.
The ERG22+ has also informed the Vulnerability Assessment Framework (VAF), a tool to assess risk of radicalisation as part of the Channel programme in the UK. The VAF uses the same items as the ERG22+ and groups them into the same categories of engagement, intent, and capability. The VAF can be used with non offenders, and is used most often with people under the age of 20 (Skleparis & Knudsen, 2020)
o the tool’s links to a treatment programme (HII);
o the lack of research proving that the factors included can predict risk, given the low base rate of extremist recidivism; and
The growing body of literature should be reviewed regularly to ensure the tool remains appropriate for use with groups such as women and young people (Lloyd, 2019).
RATED page updated: April 2021 © Risk Management Authority 2021
o the ability to add additional factors where relevant; and
o the need for more research into the tool’s validity and reliability;
Lloyd & Dean (2015) state that the tool’s basis in literature and the pathways of terrorist offenders gives the tool some empirical grounding in the absence of validation data.
o that the casework that informed the tool was mostly focused on al Qaeda inspired extremism.
Limitations include:
There is a screening version of the tool, the Extremist Risk Screen (ERS), for offenders with no previous convictions for terrorist offences. The ERS is meant to assist security staff, police liaisons, and offender managers in assessing the credibility of concerns related to potential involvement in extremism. It also informs intervention, which may include a full ERG22+ assessment (Lloyd & Dean, 2015). The ERS has good face validity and utility with assessors and offenders (Lloyd, 2019).
The ERG22+ is the intellectual property of HMPPS.
o the ability to explore pathways that do not necessarily involve violence.
Description
Tool Development
Name of Tool Multi Level Guidelines (MLG)
Category Violence Risk (Awaiting Validation)
RATED page updated: April 2021 © Risk Management Authority 2021
• MLG was developed from a literature review of group based violence, as well as feedback from experienced threat analysts (Cook et al., 2019).
• The tool lists 16 risk factors categorised into four domains in a nested model: individual, individual group, group, and group societal. These are rated for presence (Yes, Partial, or No) and relevance (Low, Medium, or High), and used to develop a formulation that includes scenarios and risk opinions that then inform case management. The tool does not include protective factors, but assessors should consider and include individualised protective factors in their formulation (Logan & Lloyd, 2015; Cook et al., 2019)
Age Appropriateness
14+ (Cook et al., 2019)
•Individuals should be reassessed as required, with a maximum of 12 months between assessments (Cook et al., 2019).
• The Multi Level Guidelines (MLG) is a 16 item Structured Professional Judgement (SPJ) approach to assessing risk of group based violence (GBV). This encompasses any violence by an individual who is aligned with, or a member of, a group. It includes terrorism as well as violence associated with gangs, organised crime, and cults. This may include lone actors who identify with, but are not a member of, a group (Cook et al., 2019)
Author / Publisher Cook, Hart and Kropp
Year 2013
• Rather than risk bandings, the output is a formulation that should communicate opinions on Future Violence/Case Prioritisation, Serious Physical Harm, and Imminent Violence (Hart et al., 2017; RTI International, 2017).
•Assessments should be completed in a team, with at least one member who is a Subject Matter Expert of the group the individual is affiliated with (Cook et al., 2019)
• An interview with the individual is encouraged, but not required. Assessors use evidence such as mental health records, corrections records,security information and intelligence, andany additional information that may be relevant (Cook et al., 2019).
Assessor Qualifications
• The tool can be used by professionals in mental health, criminal justice, or security services working with individuals who are at risk or involved in group based violence. There is no standardised training for the MLG (Cook et al., 2019; Hart et al., 2017)
•Hart et al. (2017) examined inter rater reliability for individualitems (presenceand relevance)and domains. While there was variation, the average was in the “excellent” range. Summary risk ratings were more varied: inter rater reliability regarding future violence was in the “good” range, while ratings for Serious Physical Harm and for Imminent Violence were in the “fair” range.
•Limitations included:
• The tool is in use in North America and Europe (Cook et al., 2019).
•A group of experts summarised the strengths and limitations of the tool. Strengths included:
RATED page updated: April 2021 © Risk Management Authority 2021
o the potential to miss relevant pathway offences that do not reach the threshold of terrorist violence; and
o the tool situates the individual in a social and societal/political context;
o the need for assessors to be experienced in risk assessment, as the framework is described as “lean” (Lloyd, 2019, p. 32).
•The risk factors in the Individual domain are modelled after HCR 20 V3 factors (Hart et al., 2017)
•Cook (2014)’s doctoral thesis states that the tool has face and content validity, as well as practical utility, after evaluating assessors’ confidence and knowledge gains following training and practice in using the tool.
o the individual level factors possibly being too general for a detailed assessment of terrorist risk;
General Notes
•In a doctoral thesis, Cook (2014) examined the utility and inter rater reliability of the first version of the MLG with a sample (n = 42) of assessors across 11 GBV cases. Inter rater reliability was tested for individual items (ICCs ranged from poor to excellent), domains (fair to excellent), and conclusory opinions (good to excellent). Ratings for individual items, domains, and conclusory opinions spanned all possible rating options, indicating the tool can communicate various levels of risk (Cook, 2014). The MLG subsequently underwent a revision, where four risk factors were removed (Hart et al., 2017).
o the tool can be used for various types of group based violence; and
•Hart et al. (2017) demonstrated concurrent validity in overall risk ratings between the MLG and the HCR 20, concluding that those at risk of GBV would also be identified as at risk of general violence. The developers caution that this does not indicate the reverse, and that those at risk of general violence cannot be assumed to be at risk of GBV.
o the tool’s empirical grounding that included input from subject matter experts and live practice with the tool;
o the tool demonstrated good inter rater reliability in its development.
•The MLG is open access and available to purchase without attending a standardised training course. Cook, Hart and Kropp hold the copyright for the MLG in Canada (Hart et al., 2017).
•The tool can be used in pre or post crime scenarios. It is intended for use with those at risk of committing GBV as well as those who are suspected of or known to have engaged in GBV (Hart et al., 2017).
Name of Tool Violent Extremist Risk Assessment 2 Revised (VERA 2R)
Category Violence Risk (Awaiting Validation)
Year 2016
Description
•TheVERA 2R is astructuredprofessionaljudgment(SPJ)tool that aims to assess therisk ofviolent extremism in those with histories of extremist violence or convictions for terrorism related offences. It includes 34 indicators related to violent extremism and protective factors, and an additional 11 evidence based indicators such as mental disorders and non violent criminal history (Pressman et al., 2018; Pressman et al., 2019). The latest update to the tool regarding the evidence base for risk factors was in 2018 (Pressman et al., 2019)
•The VERA 2R focuses on terrorism motivated by extremist ideologies of all types, carried out alone or within a group (Hart et al., 2017; Pressman et al., 2019). The items of the tool are divided into six domains: (1) beliefs, attitudes, and ideology; (2) social context and intention; (3) history, action, and capacity; (4) commitment and motivation; (5) protective or risk mitigating items; (6) Additional factors. Items are rated low, medium, or high, with protective and risk mitigating indicators being scored in reverse with low indicating no change, moderate indicating some positive change and high indicating asignificantpositivechange. Thefinaloutputuses risk banding as wellas anarrative formulation based on a weighting of all of the available evidence, including the findings from the risk and protective indicators (Pressman et al., 2018; Pressman et al., 2019).
•The tool is available in English, Dutch, French, and German. Finnish and Swedish versions will follow in the near future (Pressman et al., 2019).
RATED page updated: April 2021 © Risk Management Authority 2021
Author / Publisher Pressman, Rinne, Duits, and Flockton
•Writing about the VERA 2 (2012), the developers note the tool should be used alongside, and not in place of, other applicable assessments, which may include general violence risk, intellectual functioning, and personality assessments. The developers express caution when using the tool for individuals under surveillance who do not have a history of convictions for violent extremism (Pressman & Flockton, 2012; Pressman & Flockton, 2014).
•TheVERA 2R is usedin both pre crimeand post crimesituations,and can informrisk assessment, risk management, and intervention decisions (Pressman & Flockton, 2014; Pressman et al., 2019; van der Heide et al., 2019).
Age Appropriateness
Youths and adults (Pressman et al., 2019).
•Assessors integrate all known information from available sources including case file and background reports, interviews, intelligence and security assessments, and court records to determine individual item evaluative ratings. The totality of this information is considered in generating a final judgment relating to the risk of extremist violence (van der Heide et al., 2019; Pressman et al., 2019).
Assessor Qualifications
•A group of experts summarised the strengths and limitations of the VERA 2R. Strengths included:
Tool Development
o the tool’s applicability to all ideological types;
Limitationsmanagement.included:
o the inclusion of protective factors;
•Intended for use by a range of professionals trained in a variety of disciplines (security and intelligenceanalysts, forensic socialworkers, police, psychologists,and psychiatrists)in any judicial setting (forensic mental health, court, police, intelligence, prison, prosecution or any other relevant setting). It is preferred that assessors have experience in undertaking individual assessments or are professionally authorised to conduct risk assessments (Pressman et al, 2019).
o well described criteria; and
o the time and resources required;
•Herzog Evans (2018) notes that the protective factors are the inverse of six of the tool’s risk factors, and questions why the inverse of the remaining risk factors are not included as protective factors as well
•The structure of the VERA 2R is based on empirically validated SPJ tools designed to assess violence risk in adults and adolescents, including the HCR 20 (Webster et al., 1997) and SAVRY (Borum et al., 2006) (Pressman & Flockton 2014; Pressman et al., 2019).
RATED page updated: April 2021 © Risk Management Authority 2021
•Assessors must complete standardised VERA 2R training and demonstrate an understanding of the radicalisation process, violent extremism and terrorism A follow up training day every year afterwards is recommended (Pressman et al., 2019).
•Herzog Evans (2018), examining VERA 2R in a French context, notes that the tool appears not to be suitable for low level extremists, those who have not yet committed an act of violent extremism, or those who law enforcement are vaguely concerned about but for whom they lack credible evidence of extremism. Therefore the tool’s definition of violent extremist or terrorist may have a higher threshold than the definition used in some countries.
•VERA 2R is in use in Europe, North America, Australia, and Asia (Pressman et al., 2019).
General Notes
•Items on the VERA (2009) were developed based on the literature related to violent extremism. A revised and updated version, the VERA 2, was developed in 2010 following consultation with and feedback from stakeholders (Pressman & Flockton, 2012; Pressman & Flockton 2014). The VERA 2 was further updated and revised in 2016 into the VERA 2R. The current VERA 2R was most recently updated in 2018 (Pressman et al., 2019).
o that the tool provides a rich source of information for risk assessment and risk
•Beardsley & Beech (2013) applied the VERA (2009) to five case studies of high profile terrorists, using publically available information through an online search engine. The authors found that the VERA risk factors were easily applied to the case studies, across a range of ideologies and regardless of whether the individual acted alone or in a group. VERA items were relevant, thus supporting their inclusion in the tool. The inter rater reliability between two raters in the study was good, with kappa values >0.76. The authors note that a risk formulation exploring an individual’s pathway to violent extremism is necessary for proper weighting of the indicators.
•The VERA 2R is a restricted access tool only able to be purchased by individual professionals and multidisciplinary teams carrying out threat assessments, and who have undertaken the standardised training (Hart et al., 2017).
•Pressman (2016) states that the VERA 2R demonstrates deductive validity (measures factors relevant to a given conviction) when used within a legal system that uses risk factor definitions in line with those used in the tool.
• van der Heide et al., (2019) reported on a study by Van der Heide and Schuurman (2018) that found a Dutch probation service had implemented the VERA 2R but scarcely used it, citing limited practical utility due to capacity issues and lack of information Similarly, an initiative to adapt and implement the tool in Indonesiawas notsuccessful,in partdueto the resources required (Sumpter, 2020).
•Within Europe, the trademark and copyright for the VERA 2R is held by the NIFP, Dutch Custodial Services. Outside of Europe, copyright and trademark are held by D.E. Pressman.
o potential lack of access to classified information for some users; and
•There is cyber version of the VERA (CYBERA), adapted from the VERA 2 and meant to function alongside it as a complimentary tool (Gilpérez López et al., 2017).
RATED page updated: April 2021 © Risk Management Authority 2021
o the need for more empirical studies to support claims of content and deductive validity (Lloyd, 2019).
•Pressman & Flockton (2014) state that the tool demonstrated good construct validity in an unpublished study in a high security correctional facility in Australia.
• Pressman (2016) states thetool has high consumer validity, in thatexperts reported thatit assists and supports them in their assessments and professional judgements, and good face validity, in that the tool appeared relevant to users.
McClusker, P. J. (2007). Issues regarding the clinical use of the Classification of Violence Risk (COVR) assessment instrument. International Journal of Offender Therapy and Comparative Criminology, 51(6), 676 685. Access Here
McDermott, B. E., Dualan, I. V. & Scott, C. L. (2011). The predictive ability of the Classification of Violence Risk (COVR) in a forensic psychiatric hospital Psychiatric Services, 62(4), 430 433. Access Here.
M., Shaw, J., Carter, S. & Dolan, M. (2010) Investigating the validity of the classification of violence risk in a UK sample. International Journal of Forensic Mental Health, 9(4), 316 323. Access Here
Validated Tools
Monahan, J. (2010) ‘The Classification of Violence Risk.’ In R. K. Otto and K. S. Douglas. Handbook of violence risk assessment New York, London: Routledge, 187 198. Access Here
Monahan, J., Steadman, H., Robbins, P., Appelbaum, P., Banks, S., Grisso, T., Heilbrun, K., Mulvey, E. P., Roth, L. & Silver, E. (2005). An actuarial model of violence risk assessment for persons with mental disorders Psychiatric Services, 56(7), 810 815. Access Here
Monahan, J. (2001). Rethinking risk assessment: The MacArthur study of mental disorder and violence. New York: Oxford University Press. Access Here.
Kennedy, J., Bresler, S., Whitaker, H. & Masterson, B. (2007). Assessing violence risk in psychiatric inpatients: Useful tools. Psychiatric Times, 24(8). Access Here.
Fujii, D. E. M., Takioka, A. B., Lichton, A. & Hishinuma, E. (2005). Ethnic differences in prediction of violence risk with the HCR-20 among psychiatric inpatients. Psychiatric Services, 56(6), 711 716. Access Here.
Persson, M., Belfrage, H., Fredriksson, B. & Kristiansson, M (2017) Violence during imprisonment, forensic psychiatric care, and probation: Correlations and predictive validity of the risk assessment instruments COVR, LSI R, HCR 20V3, and SAPROF. International Journal of Forensic Mental Health, 16(2), 117 129. Access Here
References
Meadows, R. J. (2014) Understanding violence and victimization. New York: Pearson Education. Access Here.
Doyle,COVR
Snowden, R. J., Gray, N. S., Taylor, J. & Fitzgerald, S. (2009). Assessing risk of future violence among forensic psychiatric inpatients with the Classification of Violence Risk (COVR). Psychiatric Services, 60(11), 1522 1526. Access Here.
RATED page updated: April 2021 © Risk Management Authority 2021
Snowden, R. J., Gray, N. S. & Taylor, J. (2010) Risk Assessment for future violence in individuals from an ethnic minority group. International Journal of Forensic Mental Health, 9(2), 118 123. Access Here
de Vogel, V., de Vries Robbé, M., van Kalmthout, W. & Place, C. (2012). Risk assessment of violent women: Development of the “Female Additional Manual” (FAM) [article in Dutch].’ Tijdschrift voor Psychiatrie, 54(4), 329 – 338. Access Here.
Douglas, K. S. & Belfrage, H. (2014) Interrater reliability and concurrent validity of the HCR 20 Version 3.’ The International Journal of Forensic Mental Health, 13(2), 130 139. Access Here.
Campbell, M. A., French, S. & Gendreau, P. (2009). The prediction of violence in adult offenders: A meta analytic comparison of instruments and methods of assessment. Criminal Justice and Behavior, 36(6), 567 590. Access Here
Cawood, J. S. (2017) The inter rater reliability and predictive validity of the HCR 20 V3 in common workplace environments. Journal of Threat Assessment and Management, 4(1), 1 11. Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
M., Laix, J., Gasser, J. & Malin, V. (2014). Predicting physically violent misconduct in prison: A comparison of four risk assessment instruments. Behavioral Science and the Law, 37(1), 61 77. Access Here
Arai, K., A. Takano, T. Nagata & N. Hirabayashi. (2016) Predictive accuracy of the Historical Clinical-Risk Management 20 for violence in the forensic psychiatric wards in Japan. Criminal Behaviour and Mental Health, 27(5), 1471 2857. Access Here
Abbiati,HCR-20
Dickens, G. L. & O’Shea, L. E. (2017). Use of the HCR 20 for violence risk assessment: Views of clinicians working in a secure inpatient mental health setting.’ The Journal of Forensic Practice, 19(2), 130 138. Access Here
Douglas, K. S., Webster, C. D., Hart, S. D. & Ogloff, J. R. P. (2001) HCR 20 violence risk management guide Burnaby, Canada: Simon Fraser University. [Not accessible]
Coid, J., Yang, M., Ullrich, S., Zhang, T., Sizmur, S., Roberts, C., Farrington, D. P., & Rogers, R. D. (2009) Gender differences in structured risk assessment: Comparing the accuracy of five instruments. Journal of Consulting and Clinical Psychology, 77(2), 337 348. Access Here.
de Vogel, V. & de Ruiter, C. (2005). The HCR 20 in personality disordered female offenders: A comparison with a matched sample of males. Clinical Psychology & Psychotherapy, 12(3), 226 240. Access Here
Douglas, K. S., Guy, L. S., Reeves, K. A.; & Weir, J. (2005). HCR-20 violence risk assessment scheme: Overview and annotated bibliography Systems and Psychosocial Advances Research Center Publications and Presentations, 335. Access Here
Sturup, J., Kristiansson, M. & Lindqvist, P. (2011). Violent behaviour by general psychiatric patients in Sweden. Psychiatry Research, 188(1), 161 165. Access Here.
Doyle, M., Power, L. A., Coid, J., Kallis, C., Ullrich, S. & Shaw, J. (2014). Predicting post discharge community violence in England and Wales using the HCR-20V3 International Journal of Forensic Mental Health, 13(2), 140 147. Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
Mills, J. F., Kroner, D. G. & Hemmati, T. (2007). The validity of violence risk estimates: An issue of item performance. Psychological Services, 4(1), 1 12. Access Here
O’Shea, L. E., Thaker, D. K., Picchioni, M. M., Knight, F. L., Dickens, C. & Dickens, G. L. (2015) Predictive validity of the HCR 20 for violent and non violent sexual behaviour in a secure mental health service. Criminal Behaviour and Mental Health, 26(5), 366 379. Access Here
Lindsay, W. R., Hogue, T. E., Taylor, J. L., Steptoe, L., Mooney, P., O’Brien, G., Johnstone, S. & Smith, A. H. W. (2008) Risk assessment in offenders with intellectual disability: A comparison across three levels of security. International Journal of Offender Therapy and Comparative Criminology, 52(1), 90 111. Access Here
Green, D., Schneider, M., Griswold, H., Belfi, B., Herrera, M. & DeBlasi, A. (2016). A comparison of the HCR-20V3 among male and female insanity acquittees: A retrospective file study International Journal of Forensic Mental Health, 15(1), 48 64. Access Here.
Jeandarme, I., Pouls, C., De Laender, J., Oei, T. I. & Bogaerts, S. (2017) Field validity of the HCR-20 in forensic medium security units in Flanders. Psychology, Crime & Law, 23(4), 305 322. Access Here
Doyle, M. & Dolan, M. (2006) Predicting community violence from patients discharged from mental health services. British Journal of Psychiatry, 189(6), 520 526. Access Here.
Gray, N. S., Taylor, J. & Snowden, R. J. (2008). Predicting violent reconvictions using the HCR 20. British Journal of Psychiatry, 192(5), 384 387. Access Here
Douglas, K. S., Hart, S. D., Webster, C. D., Belfrage, H., Guy, L. S. & Wilson, C. M. (2014) Historical-Clinical-Risk Management 20, version 3 (HCR 20V3): Development and overview. International Journal of Forensic Mental Health, 13(2), 93 108. Access Here.
Garcia Mansilla, A., Rosenfeld, B. & Cruise, K. R. (2011) Violence risk assessment and women: Predictive accuracy of the HCR 20 in a civil psychiatric sample Behavioral Sciences and the Law, 29(5), 623 633. Access Here.
Douglas, K. S. & Reeves, K. A. (2010) Historical-Clinical-Risk Management 20 (HCR-20) violence risk assessment scheme: Rationale, application and empirical overview. In R. K. Otto and K. S. Douglas (Eds.), Handbook of Violence Risk Assessment. New York, London: Routledge, pp. 147 186. Access Here
Ho, H., Thomson, L. & Darjee, R. (2009). Violence risk assessment: The use of the PCL SV, HCR-20, and VRAG to predict violence in mentally disordered offenders discharged from a medium secure unit in Scotland. The Journal of Forensic Psychiatry & Psychology, 20(4), 523 541. Access Here.
Logan, C. (2014). The HCR 20 version 3: A case study in risk formulation. International Journal of Forensic Mental Health, 13(2), 172 180. Access Here
Cartwright, J. K., Desmarais, S. L., Hazel, J., Griffith, T. & Azizian, A. (2018). Predictive validity of HCR-20, START and Static 99R assessments in predicting institutional aggression among sexual offenders Law and Human Behavior, 42(1), 13 25. Access Here
Collins, M. J., Desmarais, S.L., Nicholls, T.L. & Brink, J. (2008) The Short Term Assessment of Risk and Treatability (START): Evaluating perceived utility and user satisfaction in clinical practice Poster presented at the meetings of the International Association of Forensic Mental Health Services, Vienna, Austria. [Not accessible]
Sada, A., Robles Garcia, R., Martinez Lopez, N., Hernandez Ramirez, C. A., Tovilla Zarate, F. Lopez Munguila, E., Ayala Suarez Alvarez, X. & Fresan, A. (2016). Assessing the reliability, predictive and constructive validity of historical, clinical and risk management 20 (HCR 20) in Mexican psychiatric inpatients. Nordic Journal of Psychiatry, 70(6), 456 461. Access Here
Strub, D. S., Douglas, K. S. & Nicholls, T. L. (2016). Violence risk assessment of civil psychiatric patients with the HCR 20: Does gender matter?’ International Journal of Forensic Mental Health, 15(1), 81 96. Access Here
Chu, C., Thomas, S. D. M.., Ogloff, J. R.P. & Daffern, M. (2011). The predictive validity of the short term assessment of risk and treatability (START) in a secure forensic hospital: Risk factors and strengths. International Journal of Forensic Mental Health, 10(4), 337 345. Access Here.
Alderman,START
N., Major, G. & Brooks, J. (2016). What can structured professional judgment tools contribute to the management of neurobehavioral disability?: Predictive validity of the Short Term Assessment of Risk and Treatability (START) in acquired brain injury. Neuropsychological Rehabilitation: An International Journal, 28(3), 448 465. Access Here.
Vitacco, M. J., Tabernik, H. E., Zavodny, D., Bailey, K. & Waggoner, C. (2016). Projecting risk: The importance of the HCR 20 risk management scale in predicting outcomes with forensic patients Behavioral Sciences and the Law, 34(2 3), 308 320. Access Here
Shepherd, S. M., Campbell, R. E. & Ogloff, J. R. P. (2018) The Utility of the HCR 20 in an Australian sample of forensic psychiatric patients. Psychiatry, Psychology and the Law, 25(2), 273 282. Access Here
Braithwaite, E., Charette, Y., Crocker, A. G. & Reyes, A. (2010). The predictive validity of clinical ratings of the Short Term Assessment of Risk and Treatability (START). International Journal of Forensic Mental Health, 9(4), 271 281. Access Here.
Crocker, A. G., Braithwaite, E., Laferrière, D., Gagnon, D., Venegas, C. & Jenkins, T. (2011) START changing practice: Implementing a risk assessment and management tool in a
RATED page updated: April 2021 © Risk Management Authority 2021
Schaap, G., Lammers, S. and de Vogel, V. (2009). Risk assessment in female forensic psychiatric patients: A quasi prospective study into the validity of the HCR 20 and PCL R. Journal of Forensic Psychiatry & Psychology, 20(3), 354 365. Access Here
Gray, N. S., Benson, R., Craig, R., Davies, H., Fitzgerald, S., Huckle, P., Maggs, R., Taylor, J., Trueman, M., Williams, T. & Snowden, R. J. (2011) The Short Term Assessment of Risk and Treatability (START): A prospective study of inpatient behavior. International Journal of Forensic Mental Health, 10(4), 305 313. Access Here
de Vogel, V., Bruggeman, M. & Lancel, M. (2019) Gender sensitive violence risk assessment: Predictive validity of six tools in female forensic psychiatric patients . Criminal Justice and Behavior, 46(4), 528 549. Access Here
Desmarais, S. L., Nicholls, T. L., Wilson, C. M. & Brink, J. (2012a). Using dynamic risk and protective factors to predict inpatient aggression: Reliability and validity of START assessments. Psychological Assessment, 24(3), 685 700. Access Here
civil psychiatric setting. International Journal of Forensic Mental Health, 10(1), 13 28. Access Here.
Marriott, R., O’Shea, L. E., Picchichi, M. M. & Dickens, G. L. (2017). Predictive validity of the short term assessment of risk and treatability (START) for multiple adverse outcomes: The effect of diagnosis. Psychiatry Research, 256, 435 443. Access Here.
Desmarais, S. L., Sellers, B. G., Viljoen, J. L., Cruise, K. R., Nicholls, T. L. & Dvoskin, J. A. (2012b). Pilot implementation and preliminary evaluation of START:AV assessments in secure juvenile correctional facilities International Journal of Forensic Mental Health, 11(3), 150 164. Access Here
Kroppan, E., Nesset, M. B., Nonstad, K., Pedersen, T. W., Almvik, R. & Palmstierna, T. (2011) Implementation of the Short Term Assessment of Risk and Treatability (START) in a forensic high secure unit International Journal of Forensic Mental Health, 10(1), 7 12. Access Here.
RATED page updated: April 2021 © Risk Management Authority 2021
Nicholls, T. L., Brink, J., Desmarais, S. L., Webster, C. D. & Martin, M. L. (2006). The Short Term Assessment of Risk and Treatability (START): A prospective validation study in a forensic psychiatric sample Assessment, 13(3), 313 327. Access Here
Doyle, M., Lewis, G. & Brisbane, M. (2008). Implementing the Short Term Assessment of Risk and Treatability (START) in a forensic mental health service. Psychiatric Bulletin 32, 406 408. Access Here
O'Shea, L. E. & Dickens, G. L. (2015) Predictive validity of the Short Term Assessment of Risk and Treatability (START) for aggression and self harm in a secure mental health service: Gender differences.’ International Journal of Forensic Mental Health, 14(2), 132 146. Access Here
Dickens, G. L. & O’Shea, L. E. (2015) How short should short term risk assessment be? Determining the optimum interval for START reassessment in a secure mental health service. Psychiatric and Mental Health Nursing, 122(6), 397 406. Access Here.
Kroppan, E., Nonstad, K., Iversen, R. B. & Sondenaa, E. (2017) Implementation of the Short Term Assessment of Risk and Treatability over two phases. Journal of Multidisciplinary Healthcare, 10, 321 336. Access Here
Wilson, C. M., Desmarais, S. L., Nicholls T. L. & Brink, J. (2010). The role of client strengths in assessments of violence risk using the Short Term Assessment of Risk and Treatability (START). International Journal of Forensic Mental Health, 9(4), 282 293. Access Here.
Desmarais, S. L., & Simons-Rudolph, J. (2020). Application of the TRAP 18 framework to U.S. and Western European lone actor terrorists. Studies in Conflict & Terrorism. Access Here.
Böckler, N., Hoffmann, J., & Zick, A. (2015). The Frankfurt airport attack: A case study on the radicalization of a lone actor terrorist. Journal of Threat Assessment and Management, 2(3 4), 153 163. Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
TRAP Brugh,18C.S.,
Goodwill, A., & Meloy, J. R. (2019) Visualizing the relationship among indicators for lone actor terrorist attacks: Multidimensional scaling and the validity of the TRAP 18. Behavioral Sciences and the Law [in press]
Viljoen, J. L., Beneteau, J. L., Gulbransen, E., Brodersen, E., Desmarais, S. L., Nicholls, T. L. & Cruise, K. R. (2012) Assessment of multiple risk outcomes, strengths, and change with the START:AV: A short term prospective study with adolescent offenders. International Journal of Forensic Mental Health, 11(3), 165 180. Access Here
Viljoen, S., Nicholls, T., Greaves, C., de Ruiter, C. & Brink, J. (2011). Resilience and successful community reintegration among female forensic patients: A preliminary investigation. Behavioral Science and the Law, 29(5), 752 770. Access Here
Webster, C. D., Martin, M. L., Brink, J., Nicholls, T. L. & Middleton, C. (2004) Manual for the Short Term Assessment of Risk and Treatability (START): Version 1.0 consultation Edition. Hamilton, Canada: St. Joseph's Healthcare; Port Coquitlam, Canada: Forensic Psychiatric Services Commission. [Not accessible]
Quinn, R., Miles, H. & Kinane, C. (2013). The validity of the Short Term Assessment of Risk and Treatability (START) in a UK medium secure forensic mental health service International Journal of Forensic Mental Health, 12(3), 215 224. Access Here
Challacombe, D. J., & Lucas, P. A. (2018). Postdicting violence with sovereign citizen actors: An exploratory test of the TRAP 18. Journal of Threat Assessment and Management, Advance Online Publication. Access Here
O’Shea, L. E., Picchioni, M. M. & Dickens, G. L. (2016). The predictive validity of the Short term Assessment of Risk and Treatability (START) for multiple adverse outcomes in a secure psychiatric inpatient settings. Assessment, 23(2), 150 162. Access Here.
Erlandsson, A., & Meloy, J. R. (2018). The Swedish school attack in Trollhättan. Journal of Forensic Science, 63(6), 1917 1927. Access Here.
Timmins, K. L. E., Evans, L. & Tully, R. J. (2018). Inter rater reliability of the Short Term Assessment of Risk and Treatability (START). Journal of Forensic Psychiatry and Psychology, 29(6), 968 988. Access Here.
Meloy, J. R., & Genzmas, J. (2016) The Clinical Threat Assessment of the Lone Actor Terrorist. Psychiatric Clinics of North America, 39(4), 649 662. Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
Meloy, J., R., Goodwill, A. M., Meloy, M. J., Amat, G., Martinez, M., & Morgan, M. (2019) Some TRAP 18 indicators discriminate between terrorist attacks and other subjects of national security concern. Journal of Threat Assessment and Management, Advance online publication. Access Here
M., Laix, J., Gasser, J. & Malin, V. (2014). Predicting physically violent misconduct in prison: A comparison of four risk assessment instruments. Behavioral Science and the Law, 37(1), 61 77. Access Here.
Brookstein, D., Daffern, M. & Ogloff, J. (2016) For better or worse: The predictive validity of the HCR 20 V3 and the VRAG R in community settings. International Association of Forensic Mental Health Services Conference, New York, 21 23 June 2016. [Not accessible]
Andreau Rodriguez, J. M., Peña Fernández, M. E. & Loza, W. (2016). Predicting risk of violence through a self appraisal questionnaire. European Journal of Psychology Applied to Legal Context, 8(2), 51 56. Access Here
Meloy, J. R., & Gill, P. (2016). The lone-actor terrorist and the TRAP 18. Journal of Threat Assessment and Management, 3(11), 37 52. Access Here.
Guldimann, A., & Meloy, J.R. (2020). Assessing the threat of lone actor terrorism: the reliability and validity of the TRAP 18. Forensische Psychiatrie, Psychologie, Kriminologie, 14, 158 166. Access here.
Meloy, J. R., Roshdi, L., Giaz Ocik, J., & Hoffmann, J. (2016) Investigating the individual terrorist in Europe. Threat Assessment and Management, 2(3 4), 140 152. Access Here.
Meloy, J. R. (2018). The operational development and empirical testing of the Terrorist Radicalization Assessment Protocol (TRAP 18). Journal of Personality Assessment, 100(5), 483 492. Access Here.
Meloy, J. R. (2019). Terrorist Radicalization Assessment Protocol (TRAP 18). In M Lloyd (Ed.). Extremism risk assessment: A directory. (pp. 33 38). Centre for Research and Evidence on Security Threats Access Here
Meloy, J. R., Hoffmann, J., Roshdi, K., & Guldimann, A. (2014). Some warning behaviors discriminate between school shooters and other students of concern. Journal of Threat Assessment and Management, 1(3), 203 211. Access Here
Abbiati,VRAG-R
Barra, S., Bessler, C., Landolt, M. A. & Aebi, M. (2018). Testing the validity of criminal risk assessment tools in sexually abusive youth Psychological Assessment, 30(1), 1430 1443. Access Here.
Meloy, J. R. (2016) TRAP 18 Manual. New York: Global Institute of Forensic Research. [Not accessible]
Daffern, M. (2007). The predictive validity and practical utility of structured schemes used to assess risk for aggression in psychiatric inpatient settings. Aggression and Violent Behavior, 12(1), 116 130. Access Here
Doyle, M., Carter, S., Shaw, J. & Dolan, M. (2012). Predicting community violence from patients discharged from acute mental health units in England. Social Psychiatry and Psychiatric Epidemiology, 47(4), 627 637. Access Here.
Coid, J., Yang, M., Ullrich, S., Zhang, T., Sizmur, S., Roberts, C., Farrington, D. P. & Rogers, R. D. (2009) Gender differences in structured risk assessment: Comparing the accuracy of five instruments. Journal of Consulting and Clinical Psychology, 77(2), 337 348.
Doyle, M., Dolan, M. & McGovern, J. (2002). The validity of Northern American risk assessment tools in predicting in-patient behaviour in England. Legal & Criminological Psychology, 7(2), 141 154. Access Here
Cox, J., Fairfax Columbo, J., DeMatteo, D., Vitacco, M. J., Kopkin, M. R., Parrott, C. T. & Bownes, E. (2018) An update and expansion of the role of the Violence Risk Appraisal Guide and Historical Clinical Risk Management-20 in United States case law. Behavioral Sciences and the Law, 36(5), 517 531. Access Here
Endrass, J., Rossegger, A., Frischknecht, A., Noll, T. & Urbaniok, F. (2008) Using the Violence Risk Appraisal Guide (VRAG) to predict in prison aggressive behavior in a Swiss offender population. International Journal of Offender Therapy and Comparative Criminology, 52(1), 81 89. Access Here
Camilleri, J. A. and Quinsey, V. L. (2011). Appraising the risk of sexual and violent recidivism among intellectually disabled offenders.. Psychology, Crime & Law, 17(1), 59 74.
Access Here
Harris, G. T., Rice, M.E. & Quinsey, V. L. (2016) Violence Risk Appraisal Guide Revised, 2013: User guide. Belfast, Northern Ireland: Data Services, Queen’s University Library. Access Here
Glover, A. J. J., Churcher, F. P., Gray, A. L., Mills, J. F., & Nicholson, D. E. (2017). A cross-validation of the Violence Risk Appraisal Guide Revised (VRAG R) within a correctional sample. Law and Human Behavior, 41(6), 507 518. Access Here.
Doyle, M. and Dolan, M. (2006). Predicting community violence from patients discharged from mental health services. The British Journal of Psychiatry, 189(6), 520 526. Access Here
Eisenbarth, H., Osterheider, M., Nedopil, N. & Stadtland, C. (2012). Recidivism in female offenders: PCL R lifestyle factor and VRAG show predictive validity in a German sample. Behavioral Sciences & the Law, 30(5), 575 584. Access Here
Gray, N. S., Fitzgerald, J., Taylor, J. L., MacCulloch, M. J. & Snowden, R. J. (2007). Predicting future reconviction in offenders with intellectual disabilities: The predictive efficacy of VRAG, PCL SV, and the HCR 20. Psychological Assessment, 19(4), 474 479. Access Here.
RATED page updated: April 2021 © Risk Management Authority 2021
Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
Ho, H., Thomson, L. & Darjee, R. (2009) Violence risk assessment: the use of the PCL SV, HCR-20, and VRAG to predict violence in mentally disordered offenders discharged from a medium secure unit in Scotland. The Journal of Forensic Psychiatry & Psychology, 20(4), 523 541. Access Here
Kröner, C., Stadtland, C., Eidt, M., & Nedopil, N. (2007). The validity of the Violence Risk Appraisal Guide (VRAG) in predicting criminal recidivism. Criminal Behaviour and Mental Health, 17(2), 89 100. Access Here.
Harris, G. T., Rice, M. E., Quinsey, V. L. & Cormier, C. A. (2015). Violent offenders: Appraising and managing risk (3rd ed.). Washington, DC, US: American Psychological Association. Access Here
Hertz, P. G., Eher, R., Etzler, S. & Rettenberger, M. (2019). Cross-validation of the revised version of the Violence Risk Appraisal Guide (VRAG R) in a sample of individuals convicted of sexual offences. Sexual Abuse, 1-25. Access Here
Langton, M., C., Barbaree, H., Seto, M., C., Peacock, E. J., Harkins, L. & Hansen, K. (2007). Actuarial assessment of risk for reoffense among adult sex offenders. Criminal Justice and Behavior, 34(1), 37 59. Access Here
Olver M. E. & Sewall, L. A. (2018) Cross-validation of the discrimination and calibration properties of the VRAG R in a treated sexual offender sample. Criminal Justice and Behavior, 45(6), 741 761. Access Here
Rice, M. E., Harris, G. T., Lang, C. & Chaplin, T. C. (2008). Sexual preferences and recidivism of sex offenders with mental retardation. Sexual Abuse, 20(4), 409 425. Access Here
Loza, W. & Dhaliwal, G. K. (1997). Psychometric evaluation of the Risk Appraisal Guide (RAG): A tool for assessing violent recidivism Journal of Interpersonal Violence, 12(6), 779 793. Access Here.
Pouls, C. & Jeandarme, I. (2018). Predicting institutional aggression in offenders with intellectual disabilities using the Violence Risk Appraisal Guide. Journal of Applied Research In Intellectual Disabilities, 31(2), e265 e271. Access Here.
Quinsey, L., Harris, G. T., Rice, M. E. & Cormier, C. A. (2006) Violent offenders: Appraising and managing the risk. Washington, D. C.: American Psychological Association. [Not accessible]
Rice, M. E., Harris, G. T. & Lang, C. (2013). Validation of and revision to the VRAG and SORAG: the Violence Risk Appraisal Guide Revised (VRAG R). Psychological Assessment, 25(3), 951 965. Access Here.
Hastings, M. E., Krishnan, S., Tangney, J. & Stuewig, J. (2011). Predictive and incremental validity of the Violence Risk Appraisal Guide scores with male and female jail inmates.’ Psychological Assessment, 23(1), 174 83. Access Here.
Mills, J. F., Kroner, D. G. & Hemmati, T. (2007) The validity of violence risk estimates: An issue of item performance. Psychological Services, 4(1), 1–12. Access Here.
Dolan, M., Fullam, R., Logan, C. & Davies, G. (2008). The Violence Risk Scale second edition (VRS 2) as a predictor of institutional violence in a British forensic inpatient sample. Psychiatry Research, 158(1), 55 65. Access Here
Dolan, M. & Fullam, R. (2007). The validity of the Violence Risk Scale second edition (VRS 2) in a British forensic inpatient sample. Journal of Forensic Psychiatry & Psychology, 18(3), 381 393. Access Here.
Andrews,VRS-2
Wong, S. & Gordon, A. (1999) Manual for the Violence Risk Scale. Saskatchewan, Canada: University of Saskatchewan. [Not accessible]
RATED page updated: April 2021 © Risk Management Authority 2021
Snowden, R. J., Gray, N. S., Taylor, J. & Fitzgerald, S. (2009). Assessing risk of future violence among forensic psychiatric inpatients with the Classification of Violence Risk (COVR). Psychiatric Services, 60(11), 1522 1526. Access Here.
Doyle, M., Carter, S., Shaw, J. & Dolan, M. (2012). Predicting community violence from patients discharged from acute mental health units in England. Social Psychiatry and Psychiatric Epidemiology, 47(4), 627 637. Access Here
Verbrugge, H. M., Goodman Delahunty, J. & Frize, M. C. J. (2011). Risk assessment in intellectually disabled offenders: Validation of the suggested ID supplement to the HCR-20.’ International Journal of Forensic Mental Health, 10(2), 83 91. Access Here.
Snowden, R. J., Gray, N. S. & Taylor, J. (2010). Risk assessment for future violence in individuals from an ethnic minority group in the UK. International Journal of Forensic Mental Health, 9(2), 118 123. Access Here
Rossegger, A., Laubacher, A., Moskvitin, K., Villmar, T., Palermo, G. B. & Endrass, J. (2011). Risk assessment instruments in repeat offending: The usefulness of FOTRES.’ International Journal of Offender Therapy and Comparative Criminology, 55(5), 716 731. Access Here
Stewart, C. A. (2011) Risk assessment of federal female offenders Unpublished doctoral thesis. Saskatchewan, Canada: University of Saskatchewan. Access Here.
Thomson, L., Davidson, M., Brett, C., Steele, J. & Darjee, R. (2008) risk assessment in forensic patients with schizophrenia: The predictive validity of actuarial scales and symptom severity for offending and violence over 8 – 10 years. International Journal of Forensic Mental Health, 7(2), 173 189. Access Here
D. A. & Bonta, J. (2010). The Psychology of Criminal Conduct (5th edition). Salisbury, UK: Anderson Publishing Ltd. Access Here.
Daffern, M. (2007). The predictive validity and practical utility of structured schemes used to assess risk for aggression in psychiatric inpatient settings. Aggression and Violent Behavior 12(1), 116 130. Access Here.
Lewis, K., Olver, M. E. & P., S. C. (2013). The Violence Risk Scale: Predictive validity and linking changes in risk with violent recidivism in a sample of high- risk offenders with psychopathic traits Assessment, 20(2), 150 164. Access Here.
Yesberg, J. A., Scanlan, J. M. & Polaschek, D. L. L. (2014). Women on parole: do they need their own DRAOR? Practice : The New Zealand Corrections Journal, 2(1), 20 25. Access Here.
Chadwick, N. (2014). Validating the Dynamic Risk Assessment for Offender Re entry (DRAOR) in a sample of U.S. probationers and parolees [Master’s thesis, Carleton University]. Carleton University Research Virtual Environment (CURVE) Access Here
Wong, S. C. P. & Gordon, A. (2006). The validity and reliability of the Violence Risk Scale: A treatment friendly violence risk assessment tool. Psychology, Public Policy, and Law, 12(3), 279 309. Access Here.
Wong, S. C. P & Parhar, K. K. (2011). Evaluation of the predictive validity of the Violence Risk Scale in a paroled offender sample: a seven year prospective study. The Journal of Forensic Psychiatry & Psychology, 22(6), 790 808. Access Here
Zhang X. L., Chen X. C., Cai W. X. & Hu, J. M. (2012). Reliability of the violence risk scale of Chinese version [article in Chinese] Fa Yi Xue Za Zh, 28(1), 32 35. Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
Tamatea, A. & Wilson, N. (2009). Dynamic Risk Assessment for Offender Re entry (DRAOR): A pilot study. Wellington, New Zealand: New Zealand Department of Corrections. [Not accessible]
A. E. (2016) An investigation of the dynamic risk assessment for offender re entry (DRAOR) with New Zealand sexual offenders [Master’s thesis, University of Canterbury]. University of Canterbury Research Repository. Access Here.
Tools awaiting validation
Serin, R. C., Lloyd, C. D. & Hanby, L. J. (2010). Enhancing offender re entry an integrated model for enhancing offender re entry. European Journal of Probation, 2(2), 53 75. Access Here
Hanby, L. (2013) A longitudinal study of dynamic risk, protective factors and criminal recidivism: change over time and the impact of assessment timing [Doctoral thesis, Carleton University]. Carleton University Research Virtual Environment (CURVE). Access Here
Yesberg, J. A. & Polaschek, D. L. L. (2014). What can the DRAOR tell us about high risk offenders? A preliminary examination. Practice: The New Zealand Corrections Journal, 2(1), 13 19. Access Here.
Polaschek, D. L. L. & Yesberg, J. A. (2018). High-risk violent prisoners’ patterns of change on parole on the DRAOR’s dynamic risk and protective factors. Criminal Justice and Behavior, 45(3), 340 363. Access Here
Perley Robertson, B. (2019). Predictive validity of the dynamic risk assessment for offender re entry among intimate partner violence offenders. [Master’s thesis, Carleton University]. Carleton University Research Virtual Environment (CURVE). Access Here.
Averill,DRAOR
(2014). The healthy identity intervention: The UK’s development of a psychologically informed intervention to address extremist offending. In M. Lloyd (Ed.). Extremism risk assessment: A directory. (pp. 89 107). Routledge. Access Here.
Herzog Evans, M. (2018). A comparison of two structured professional judgment tools for violent extremism and their relevance in the French context. European Journal of Probation, 10(1), 3-27. Access Here
Cook,MLG
Dean,ERG22+C.
Skleparis, D., & Knudsen, R.A. (2020). Localising ‘radicalisation’: Risk assessment practices in Greece and the United Kingdom. British Journal of Politics and International Relations, 22(2), 309 – 327. Access here.
HMPPS. (2019). Extremism risk guidelines: ERG22+. In M. Lloyd (Ed.), Extremism risk assessment: A directory (pp. 12 18). Centre for Research and Evidence on Security Threats, Access Here.
Lloyd, M. (2019). Extremism risk assessment: A directory. Centre for Research and Evidence on Security Threats. Access Here.
Van der Heide, L., van der Zwan, & van Leyenhorst, M. (2019), ‘The practitioner’s guide to the galaxy – A comparison of risk assessment tools for violent extremism’, ICCT research paper (Online) Access Here
RATED page updated: April 2021 © Risk Management Authority 2021
Logan, C., & Lloyd, M. (2018). Violent extremism: A comparison of approaches to assessing and managing risk. Legal and Criminal Psychology, 24(1), 141 161. Access Here.
Powis, B., Randhawa Horne, K., & Bishopp, D. (2019). The structural properties of the extremism risk guidelines (ERG22+): A structured formulation tool for extremist offenders. Ministry of Justice Analytical Series. Access Here.
Yesberg, J. A., Scanlan, J. M., Hanby, L. J., Serin, R. C. & Polaschek, D. L. L. (2015). Predicting women’s recidivism: validating in dynamic community based “gender neutral” tool. Probation Journal, 62(1), 33 48. Access Here.
Dean, C., Lloyd, M., Keane, C., Powis, B., & Randhawa, K. (2019). Intervening with extremist offenders – A pilot study. HM Prison & Probation Service. Access Here.
A.N., Hart, S.D., & Kropp, P.R (2019). ‘Multi level guidelines: MLG version 2’, in Lloyd, M. (ed.) Extremism risk assessment: A directory (pp. 28 32) Centre for Research and Evidence on Security Threats, 28 – 32 Access Here.
Hart, S. D., Cook, A. N., Pressman, D. E., Strang, S. & Lim, Y. L. (2017) A concurrent evaluation of threat assessment tools for the individual assessment of terrorism. Ontario, Canada: The Canadian Network for Research on Terrorism, Security and Society, working paper series no 17- 1. Access Here.
Lloyd, M., & Dean, C. (2015). The development of structured guidelines for assessing risk in extremist offenders. Journal of Threat Assessment and Management, 2(1), 41 52. Access here.
Powis, B., Randhawa Horne, K., & Elliott, I. (2019). Inter rater reliability of the extremism risk guidelines 22+ (ERG 22+). Ministry of Justice Analytical Series. Access Here.
de Vries Robbé, M. & de Vogel, V. (2012). SAPROF 2nd edition Manual: updated Research chapter. Utrecht: Van der Hoeven Kliniek. Access Here
de Vries Robbé, M., de Vogel, V. & Douglas, K. S. (2013) Risk factors and protective factors: a two sided dynamic approach to violence risk assessment. The Journal of Forensic Psychiatry and Psychology, 24(4), 440 257. Access Here
Cook, A.N. (2014). Risk Assessment and Management of Group Based Violence [PhD dissertation, Simon Fraser University]. https://summit.sfu.ca/item/14289
RTI International (2017). Countering violent extremism - Developing a research roadmap. Access Here.
RATED page updated: April 2021 © Risk Management Authority 2021
Lloyd, M. (2019). Extremism risk assessment: A directory. Lancaster, UK: Centre for Research and Evidence on Security Threats. Access Here.
Logan, C., & Lloyd, M. (2019). Violent extremism: A comparison of approaches to assessing and managing risk. Legal and Criminal Psychology, 24(1), 141 161. https://doi.org/10.1111/lcrp.12140
Abidin, Z., Davoren, M., Naughtor, L., Gibbons, O., Nulty, A. & Kennedy, H. G. (2013) Susceptibility (risk and protective) factors for in-patient violence and self harm: Prospective study of structured professional judgment instruments: START and SAPROF, DUNDRUM 3 and DUNDRUM 4 in forensic mental health services. BMC Psychiatry, 13, 197 214. Access Here.
Hart, S. D., Cook, A. N., Pressman, D. E., Strang, S. & Lim, Y. L. (2017). A concurrent evaluation of threat assessment tools for the individual assessment of terrorism. Ontario, Canada: The Canadian Network for Research on Terrorism, Security and Society, working paper series no 17- 1. Access Here.
Coid, J. W., Kallis, C., Doyle, M., Shaw, J. & Ullrich, S. (2015). Identifying causal risk factors for violence among discharged patients. PLOS One, 10(11), e0142493. Access Here
Abbiati,SAPROF
de Vogel, V., de Vries Robbe, M., de Ruiter, C. & Bauman, Y. H. A. (2011). Assessing protective factors in forensic psychiatric practice: Introducing the SAPROF.’ International Journal of Forensic Mental Health, 10(3), 171 177. Access Here
de Vogel, V., de Ruiter, C., de Bouman, Y. & de Vries Robbé, M. (2009). SAPROF. Guidelines for the assessment of protective factors for violence risk [English version of the Dutch original]. English Version. Utrecht, The Netherlands: Forum Educatief. [Not accessible]
M., Laix, J., Gasser, J. & Malin, V. (2014). Predicting physically violent misconduct in prison: A comparison of four risk assessment instruments. Behavioral Science and the Law, 37(1), 61 77. Access Here.
de Vries Robbé, M., de Vogel, V. & de Spa, E. (2011). Protective factors for violence risk in forensic psychiatric patients: A retrospective validation study of the SAPROF.’ International Journal of Forensic Mental Health, 10(3): 178 186. Access Here.
Hart, S. D., Cook, A. N., Pressman, D. E., Strang, S. & Lim, Y. L. (2017) A concurrent evaluation of threat assessment tools for the individual assessment of terrorism. Ontario, Canada: The Canadian Network for Research on Terrorism, Security and Society, working paper series no 17- 1. Access Here.
Herzog Evans, M. (2018). A comparison of two structured professional judgment tools for violent extremism and their relevance in the French context. European Journal of Probation, 10(1), pp. 3 27. Access Here
Lloyd, M. (2019) Extremism risk assessment: A directory. Lancaster, UK: Centre for Research and Evidence on Security Threats. Access Here.
de Vries Robbe, M. & Willis, G. M. (2017). Assessment of protective factors in clinical practice.’ Aggression and Violent Behaviour, 32, 55 63. Access Here
Yoon, D., Spehr, A. & Briken, P. (2011) Structured assessment of protective factors: a German pilot study in sex offenders. The Journal of Forensic Psychiatry & Psychology, 22(6), 834 844. Access Here.
Kashiwagi, H., Kikuchi, A., Koyama, M., Saito, D. & Hirabayashi, N. (2018) Strength based assessment for future violence risk: A retrospective validation study of the Structured Assessment of Protective Factors for Violence Risk (SAPROF) Japanese version in forensic psychiatric inpatients. Annals of General Psychiatry, 17(5). Access Here
de Vries Robbé, M., de Vogel, V., Koster, K. & Bogaerts, S. (2015). Assessing protective factors for sexually violent offending with the SAPROF.’ Sexual Abuse: A Journal of Research and Treatment, 27(1), 51 70. Access Here
Pressman, D. E. (2009) Risk assessment decisions for violent political extremism. Ottawa, Canada: Public Safety Canada, Government of Canada. Access Here.
Yoon, D., Turner, D., Klein, V., Rettenberger, M., Eher, R. & Briken, P. (2016). Factors predicting desistance from reoffending: A validation study of the SAPROF in sexual offenders. International Journal of Offender Therapy and Comparative Criminology, 62(3), 697 716. Access Here.
Pressman, D.E. (2016). ‘The complex dynamic causality of violent extremism: Applications of the VERA 2 risk assessment method to CVE initiatives’, in Masys, A. (ed.) Disaster f orensics. Advanced sciences and technologies for security applications. Springer, Cha m, 249 269. Access Here.
VERA Beardsley,2R
RATED page updated: April 2021 © Risk Management Authority 2021
N. L., & Beech, A. R. (2013). Applying the violent extremist risk assessment (VERA) to a sample of terrorist case studies. Journal of Aggression, Conflict, and Peace Research, 5(10), 4 15. Access Here
Gilpérez López, I., Torregrosa, J., Barhamgi, M., & Camacho, D. (2017). An initial study on radicalisation risk factors: Towards an assessment software tool. 28th International Workshop on Database and Expert Systems Applications. Access Here.
RATED page updated: April 2021 © Risk Management Authority 2021
Pressman, D. E. & Flockton, J. (2012). Calibrating risk for violent political extremists and terrorists: The VERA 2 structured assessment. The British Journal of Forensic Practice, 14(4), pp. 237 251. Access Here
Pressman, D. E. & Flockton, J. (2014). Violent extremist risk assessment: Issues and applications of the VERA 2 in a high-security correctional setting. in A. Silke (ed.) Prisons, terrorism and extremism: Critical issues in management, radicalisation a nd reform. London, United Kingdom: Routledge, pp. 122 143. Access Here.
E., Duits, N., Rinne, T. & Flockton, J. S. (2019) ‘Violent extremism risk assessment version 2 revised (VERA 2R)’, in Lloyd, M. (ed.) Extremism risk assessment: A directory. Lancaster, UK: Centre for Research and Evidence on Security Threats, 39 45. Access Here.
WAVR 21
Sumpter, C. (2020) ‘Realising violent extremist risk assessments in Indonesia: Simplify and collaborate’. Journal for Deradicalization, Spring(22), pp. 97 – 121. Access Here
Meloy, J. R., White, S. G. & Hart, S. (2013). Workplace assessment of targeted violence risk: the development and reliability of the WAVR 21. Journal of Forensic Sciences, 58(5), 1353 1358. Access Here
Pressman, D. E., Duits, N., Rinne, T. & Flockton, J. S. (2018). Violent extremism risk assessment, version 2 revised (VERA 2R). Radicalisation Awareness Network. Access Pressman,Here.D.
Kienlen, Kristine. (n.d.). Review of the WAVR 21: A structured professional guide for the Workplace Assessment of Violence Risk (second edition). National Institute for the Prevention of Workplace Violence. Access Here.
Van der Heide, L., van der Zwan, & van Leyenhorst, M. (2019), ‘The practitioner’s guide to the galaxy – A comparison of risk assessment tools for violent extremism’, ICCT research paper (Online). Access Here
Brunt, B. V. (2013). A comparative analysis of threat and risk assessment measures. The Journal of Campus Behavioral Intervention, 1, 111 151. Access Here
Description
Assessor Qualifications
RATED page updated: July 2019 © Risk Management Authority 2019
•The purpose of the EARL 20B is to assess risk and assist in the development of risk management plans that may counteract future offending and anti social behaviour of high risk boys.
Author / Publisher Augimeri and Colleagues
Strengths
Age Appropriateness
For boys ages 12 and under
• The tool has been translated into translated into six languages other than English (Swedish, Finnish, Norwegian, French, Dutch, and Japanese) and is adapted for use with other offending populations (i.e. Maori population in New Zealand).
The EARL 20B should be used by clinicians and other professionals experienced in working with high risk children.
•The EARL 20B is 20 item structured clinical risk assessment tool developed for use with boys aged 12 and under. In clinical settings, the age range falls between 6 and 11 years of age (Augimeri et al., 1998; 2001; 2011).
Category Youth Assessment: General Risk (Validated)
Empirical Grounding
•The EARL 20B is modelled on the structure and content of the HCR 20 (Webster, et al. 1997).
Year 2001
Name of Tool Early Assessment Risk List for Boys (EARL 20B)
•Assessors have the opportunity to assign an overall clinical judgement rating of ‘low,’ ‘moderate’ or ‘high’ risk (Augimeri et al., 1998; 2001; 2019).
•Items are categorised under three sections: (1) Family, (2) Child and (3) Responsivity. The ‘Family’ items assess the nature of familial support and other environmental factors (e.g. neighbourhood). The ‘Child’ items assess individual risk factors associated with the child. The ‘Responsivity’ items assess the ability and willingness of both the family and child to engage in services. Items are rated on a 3 point scale from 0 for not present, 1 for some presence and 2 for those with a clear presence. There is a ‘clinical risk’ column allowing clinicians to apply red flags to factors of particular concern
RATED page updated: July 2019 © Risk Management Authority 2019
•Augimeri et al. (2009) using cox regression analysis, it was found that the composite scores were significantly related to an increased probability to engage in future criminal offences.
Inter-Rater Reliability
•Enebrink et al. (2006a) the EARL 20B total scores achieved moderate to large pearson correlation coefficients for proactive and reactive aggression at the 6 and 30 month follow up periods compared to clinical judgement scores which were largely non significant.
b)International Research
•Enebrink et al. (2006b) looked at a Swedish translation of the EARL 20B, which has some minor adjustments to the original. Poor IRR was obtained for Abuse/Neglect/Trauma and Coping Ability items (kappas of 0.30 and 0.38 respectively). The authors surmise that this could be the result of items being operationalised in a broad manner. The EARL 20B composite score achieved excellent inter rater reliability (ICC = .92).
a)UK Research
•de Ruiter and Augimeri (2012) the EARL 20B attained strong predictive accuracy between the composite scores (AUC =.77) and final risk judgement (.77) and delinquency reported by teachers. The EARL 20B composite scores also had moderate accuracy in predicting general (AUC = .62) and violent recidivism (AUC = .69) as documented in official police records.
General Predictive Accuracy
•Hrynkiw Augimeri (2005) the EARL 20B composite score achieved an ICC value of .82 from a review of 100 common files. The three subscales attained moderate to high ICC values ranging between .55 and .79.
•The EARL 20B is based on research relating to child development and delinquency. Author related peer reviewed studies have shown the EARL assessments to be static and dynamic tools with its component factors having sound empirical grounding (Augimeri et al., 2011).
•Developed from adult assessment tools and juvenile offending screening assessments (Augimeri et al., 2005).
•Augimeri et al. (2001) the composite EARL 20B scores attained a high ICC value of .80.
a)UK Research
Validation History
•Koegl (2011) the composite EARL 20B score achieved a moderate AUC value of .66 for any conviction.
•Enebrink et al. (2006a) found significant moderate to large correlation coefficients between the composite score and the total scores of reactive (hostile or affective) and proactive (goal orientated, instrumental or predatory) aggression in a 6 month follow up period (r = .31 and .53 respectively). At the 30 month follow up period, these correlations had decreased to .20 (ns) for reactive aggression.
•de Ruiter and van Domburgh (2016) found through a Receiver Operating Characteristic (ROC) analysis that all EARL scales significantly predicted self reported delinquency at 1 and 2 year follow up, a Disruptive Behavior Disorder (DBD) at 2 year follow up (range AUC = .70 to .79), new police registrations (range AUCs= .58 to .61), and new police registrations for violent offending (range AUCs = .59 to .69).
•Hrynkiw Augimeri (2005) the mean composite EARL 20B scores were significantly higher for boys who were found guilty of an offence than for boys who were not found guilty.
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
The EARL 21G (Levene et al. 2001) has been developed for use with high risk girls under the age of 12 please refer to the ‘Responsivity’ section of RATED.
Applicability: Females
•In a sample of 573 boys, several items on EARL 20B were found to predict risk: Caregiver Continuity, Parenting Style, Onset of Behavioural Difficulties, Likeability, Peer Socialization, Authority Contact, Antisocial Attitudes, Antisocial Behaviour and Family Responsivity (Augimeri et al., 2010a).
•In a longitudinal study with 379 boys, EARL 20B total scores significantly predicted conviction status between 15 20 years later (police records) for total offending (AUC = .64) and for three offence subtypes (i.e., property, person, administration of justice) with AUC values ranging between .60 and .63. The strongest predictors for being convicted for any offence were Antisocial Attitudes (OR = 2.64) and low Child Responsivity (OR = 2.20 (Koegl, Farrington and Augimeri, 2019).
Validation History
Applicability: Ethnic Minorities
•The EARL 20B was used in Edinburgh and is currently being used in Glasgow as part of SNAP® (Stop Now And Plan) pilot programs. The EARL 20B has also been used in New Zealand with the Mauri population and in the United States with both African American and Hispanic children
•Other studies have used factor analysis methodology to validate the underlying constructs relating to the tool (e.g., confirmatory factor analysis).
Validation History
Validation History
Applicability: Mental Disorders
No empirical evidence available.
•Koegl (2011) found significant differences in relation to the costs of custodial and probation dispositions as a function of the three clinical risk judgement categories. Boys who were rated as ‘high’ and ‘moderate’ risk incurred the highest costs in comparison to boys classified as ‘low’ risk.
•No validation evidence for UK samples.
•Despite the fact that the assessment has been translated into various languages, there have been no studies looking at the tool’s predictive validity in various ethnic groups.
•The EARL 20B does not have a single algorithm for assessing low, medium or high risk levels. Conversely, the final estimate of risk is calculated by weighing EARL 20B items and possible risk or protective factors on a case specific basis (Enebrink et al. 2006b).
No empirical evidence available.
•The majority of the current validation literature has been conducted by the authors of the EARL 20B.
•The tool can contribute to the measurement of progress / deterioration in factors related to the individual’s offending behaviours.
Contribution to Risk Practice
•The EARLs were developed in an applied accredited children’s mental health centre initially as part of the assessment process for the evidence based SNAP model. The majority of SNAP children experience clinical levels of internalising and externalising behaviour problems associated with mental health issues (e.g., disruptive behaviour problems such as conduct and oppositional disorder; Augimeri et al., 2017; 2018)
RATED page updated: July 2019 © Risk Management Authority 2019
•The EARL 20B can aid assessors in identifying risk and responsivity factors specific to the individual’s offending behaviours. These factors can also act as targets for change.
Other Considerations
•The EARL 20B can contribute to the formulation of offence analyses and risk management plans.
•The authors caution about the use of cut off scores to make decisions about a boy’s risk potential.
•The EARL Pre Checklist (EARL PC: Augimeri et al., 2010b) has been recently developed as an abbreviated version of the full EARL assessments.
•Fewer studies examining the predictive accuracy of the final judgement ratings.
RATED page updated: July 2019 © Risk Management Authority 2019
•The EARL PC is designed to screen for risk factors in children that pose a potential risk of engaging in future antisocial behaviours. It was created in response to the need for a simpler, condensed version of the EARL for use by professionals working in the criminal justice and educational sectors in cases where it may not be feasible to administer a full assessment. There are no validation data available on the EARL PC at present.
RATED page updated: July 2019 © Risk Management Authority 2019
Assessor Qualifications
•The YASI includes both pre screen and full assessment components and is used to assist in making initial service decisions as well as case plan development. Youth are rated as low, medium or high risk to reoffend.
Category Youth Assessment: General Risk (Validated)
•The instrument is used in a variety of juvenile justice settings with both males and females. A special version of the instrument is available for high risk youth serving custody sentences for serious offenses.
Strengths
•The full YASI assessment consists of 90 items spread across 10 subscales. It takes between 30 to 60 minutes to administer.
Name of Tool Youth Assessment and Screening Instrument (YASI)
•YASI provides a graphic profile of risk, need, and strength results for each youth including overall static and dynamic scores on risk and protective factors. These items are spread across 10 domains: legal history, family, school, community and peers, alcohol and drugs, mental health, aggression, prosocial and antisocial attitudes, social and cognitive skills and employment/free time.
12 18
Author / Publisher Orbis Partners Inc.
•YASI is completed by juvenile justice case workers after a file review, interview with the youth and family (where possible), and consultation with other relevant collateral sources.
Age Appropriateness
Assessors must undertake the necessary training in order to administer this tool.
•The screening version of the tool (‘pre screen’) contains 31 items, and is used to identify moderate to high risk youth who require more extensive assessment using the Full Assessment. This takes around 20 to 40 minutes to complete.
•The tool includes items pertaining to mental health, including adverse childhood experiences.
•The YASI has been modified for local legal terminology and used with youth in Scotland.
•A pre screen version is available for planning and triage purposes, with a pre screen risk score totalled from 33 items (Scott, Brown and Skilling, 2019).
•Also included are evaluation of strengths, which are assigned numerical weights.
Description
•The suggested interview questions are tailored towards the young person being assessed; although these can be adapted for an interview with their parents or to suit the particular circumstances of the young person (Baird et al., 2013).
Year 2007
a)UK Research
Empirical Grounding
•Scott, Brown and Skilling (2019) applied the YASI to 254 justice involved youth (148 males, 106 females). Using a subsample of twenty cases, good or excellent IRR was generated for each subdomain of the YASI.
•The YASI is predominantly grounded in the General Personality and Cognitive Social Learning Theory (GPCSL), with eight central factors based on social learning and self control theory. Nine out of ten of the YASI’s global subdomains pertain to the Central Eight (Scott, Brown and Skilling, 2019).
b)International Research
The YASI is a modified version of the Washington State Juvenile Court Assessment (WSJCA; Barnoski, 2004). Jones et al. (2015) states that the quantitative inclusion of strengths is apt to enhance the functions of prediction and case management. Moreover, it is claimed that the identification of a buffering effect of strengths on risk supports a critical elements of the overall YASI assessment model.
General Predictive Accuracy
None available at present.
•With a sample of 1919 juveniles on probation, Baird et al. (2013) found an average scoring agreement among 76 probation staff raters approaching .89.
Validation History
•In Illinois, over a one year period, Orbis Partners (2007) found AUCs of .65.
•Although the YASI is primarily grounded in gender neutral literature, it features a number of gender responsive items extrapolated from feminist literature (Jones et al., 2016; Scott, Brown and Skilling, 2019).
a)UK Research
None available at present.
b)International Research
•As reported in Orbis Partners Inc.’s (2018) evaluation of existing research, an inter rater reliability study was conducted with 76 raters across ten case studies. The scoring agreement amongst juvenile probation staff was 85%; whilst raters of agreement between staff and expert raters was around 80%.
•Over a two year period in New York State, an AUC of .65 was found by Orbis Partners in 2007.
RATED page updated: July 2019 © Risk Management Authority 2019
Inter Rater Reliability
•Looking at a sample of 254 youth from Ontario, it was found that the YASI pre screen yielded an AUC value for the risk total of .65; whilst the total protective score was .55. In terms of the full YASI assessment, the total score, total risk score and total protective scores generated AUCs of .66, .65 and .64 respectively (Scott, Brown and Skilling, 2019).
Applicability: Females
•Baird et al. (2013) reported an AUC of .68 for predictive validity.
•Orbis Partners Inc. (2018) reported that a new YASI validation study was carried out in Milwaukee County on 2712 youth. An AUC value was 0.76 was achieved and this increased as follow up extended from 12 to 36 months. Predictive accuracy was also evident for girls and boys, those aged under 12 years old and different ethnic groups.
•In Alberta (Canada), an AUC of .79 was reported. The high predictive risk and strength domains were Legal History (AUC Risk .73), Community and Peers (AUC Risk .72, AUC Strength .67) and Attitudes (AUC Risk .69, AUC Strength .69) (Jones, 2013; Jones et al., 2014).
Validation History
In 2007, Orbis Partners developed YASI G as a response to the assessment needs of young females. This consists of items extrapolated from feminist and gender responsive literature about female criminality: nature of one’s relationships, level of emotional expression, self efficacy, sexual vulnerability, early parenthood and potential mental health issues (Jones et al., 2016). Scott, Brown and Skilling (2019) suggested that its inclusion of gender neutral and gender responsive items means the YASI may be a particularly good choice for use with justice involved females.
None available at present.
b)International Research
•Data indicated that juvenile females were being over classified by YASI Pre Screen scores in that high risk girls exhibited lower recidivism than high risk boys (Orbis Partners Inc., 2007).
•After the initial data indicated that females were being over classified (in that high risk girls were generated a lower recidivism rate than high risk boys), separate cut off points were devised to address the gender scoring differences (Baird et al., 2013).
a)UK Research
RATED page updated: July 2019 © Risk Management Authority 2019
None available at present.
•A higher score was found for Aboriginal versus non Aboriginal individuals in Alberta Canada (Jones et al., 2014).
Applicability: Mental Disorders
•Jones et al. (2016) found there was a moderate degree of predictive accuracy in predicting general reoffending for girls with an AUC of .68 compared to the high levels of accuracy for males yielding an AUC of .82.
Applicability: Ethnic Minorities
•When using the YASI, Baird et al. (2013) found there was moderate discrimination between Whites and Blacks/African Americans. The recidivism rate, however, was only 2.7% higher for high risk Whites than moderate risk Blacks.
No empirical evidence at present.
Validation History
•A study compared the results of the YASI on 148 males with 106 females in Canada. Results indicated that the pre screen yielded higher predictive accuracy for males. The risk total and total protective scores were .68 and .61 for males and .62 and .52 for females. For the full assessment, moderate effects were observed for males. The total score, total risk score and total protective score were .70, .69 and .69 respectively for males. This is in comparison to .64, .62 and .62 for total, total risk and total protective scores or females (Scott, Brown and Skilling, 2019).
Validation History
b)International Research
a)UK Research
RATED page updated: July 2019 © Risk Management Authority 2019
•AUC’s of .78 and .76 for girls and boys respectively in the Milwaukee County study (Orbis Partners Inc., 2018).
•Robinson and Jones (2017) found that predictive accuracy levels were similar across various ethnic groups, with an AUC of .76 for African Americans, .79 for Caucasians and .73 for Hispanic.
•The modification of case plans is supported by the use of YASI in monitoring supervision progress.
•Scott, Brown and Skilling (2019) found the YASI had strong convergent validity with the YLS/CMI.
•Assessment and re assessment over a short period (up to six months) has shown risk levels relate to the presence/absence of protective factors.
RATED page updated: July 2019 © Risk Management Authority 2019
•A separate version of the instrument (CA YASI) has been developed and contains more items. It is geared toward more violent youth and is used with up to 25 years old. For this version, Skeem et al. (2012) reported ICC scores between .51 to .72 from field staff across different sites in California. The original CA version was replaced in 2017 with a streamlined CA YASI to help increase reliability and reduce the number of assessment items and is now being used with other high risk custody populations.
Contribution to Risk Practice
•Developing an understanding of strengths is appropriate to assessment and service.
Other Considerations
•Provides a profile of criminogenic needs. The authors of the tool caution that although the YLS/CMI can act as an aid to case management and planning, it is not designed to replace professional judgment.
Strengths
•The authors note that in some circumstances, the assessor might feel that the level of risk/need is different from that provided by the YLS/CMI because of factors that are not represented in the ratings. In those situations, a ‘professional override’ measure might be used. This feature allows the assessor to provide their own risk level estimate based on the information they hold about the individual
RATED page updated: July 2019 © Risk Management Authority 2019
•The YLS/CMI 2.0 is a 42 item standardised inventory for use with male and female juveniles to assess the risk of future offending. The original YLS/CMI was used in on juvenile probationers in Canada and was updated to a 2.0 version using a normative sample of 17, 000 young individuals who had offended in 2011. The revised version expanded the age range to 12 18 years, added more non criminogenic needs and responsivity considerations, and included new recommended cut offs for risk/need levels.
Age Appropriateness
•Scoring of risk factors provides an estimate of the risk of reconviction for individuals over a 12 month period. Risk levels are classified as low, moderate, high or very high.
Year 2011
•It assesses eight categories of risk factors associated with recidivism and need factors that assist in case management. Space to record narratives is also included to allow the assessor to record information like special circumstances that are not captured in the risk and needs factor items. Protective factors for the young person are also documented.
Author / Publisher Hoge and Andrews
Category Youth Assessment: General Risk (Validated)
Assessors should possess training and experience in youth assessment.
•The tool incorporates a case management section which can aid case planning and management. Its purpose is twofold in nature: recidivism predictions; addressing programming and service needs
Description
Name of Tool Youth Level of Service/Case Management Inventory 2.0 (YLS/CMI 2.0)
•Meta analyses of previous empirical research on the YLS/CMI indicated that it is useful in predicting recidivism in both males and females (Pusch and Holtfreter, 2018).
12 18
Assessor Qualifications
•Can be used by a variety of professionals with the relevant training to administer and score the assessment.
•The tool is derived from LSI R. Studies examining the psychometric properties of the YLS/CMI are presented in the manual (Hoge and Andrews, 2003).
(Anderson et al., 2016). For example, if a young person scores as high risk on the education subscale, this denotes that additional services should target this area (Barnes et al , 2016).
•The empirical evidence for the YLS/CMI 2.0 has been derived from the LSI R.
Empirical Grounding
•The authors of the YLS/CMI maintain that the tool fits the ‘Risk Needs Responsivity’ model (see Andrews et al., 2011), by giving an insight into the risks and needs of individuals and choosing the most appropriate treatment options for them.
•Rennie and Dolan (2010) found excellent inter rater reliability for the YLS/CMI (ICC =.95).
•Andrew and Bonta’s (1994) ‘Psychology of Criminal Conduct’ framework advances individual personality items and social circumstances considered to be indicative of recidivism. These factors are known as the ‘Central Eight’ and form the basis of the eight items measured in the YLS/CMI (Cuervo and Vilanueva 2018).
•Schmidt et al. (2006) found moderate to large inter rater reliability values for the separate subscales of the YLS/CMI.
a)UK Research
•Can be less time consuming to complete than other risk assessments such as the ASSET (Burman et al., 2007).
b)International Research
Inter Rater Reliability
•Vaswani (2013) found it was a good predictor of reoffending for both males and females. The authors explained that education, employment, family and circumstances/planning items may be scored as female specific factors to aid with case management (Hoge and Andrews, 2011).
•Measures dynamic variables as well as static ones. This allows for the assessment of change in risk level and also informs intervention needs and targets (Yates, 2005).
•The tool considers vulnerability, care and risk of harm factors, such as marital conflict within the family, poor social skills, victim of bullying, etc.
•Vieira et al (2009) reported very high (.98) inter rater reliability.
RATED page updated: July 2019 © Risk Management Authority 2019
•The YLS/CMI was further developed on the basis of consultation with experienced probation officers and juvenile justice professionals in order to ensure the utility of the measure (Hoge and Andrews, 2003).
•Onifade et al. (2008a) found excellent levels of agreement in the scoring of this instrument (90%). Following training, the inter rater reliability exceeded 90%.
a)UK Research
RATED page updated: July 2019 © Risk Management Authority 2019
•Welsh et al. (2008) found good ICC value for the YLS/CMI composite score (.72). ranging from .61 for the peer relations subscale to .85 for education and employment subscale.
General Predictive Accuracy
Validation History
•Rennie and Dolan (2010) examined males in England and found that the instrument gave significant predictions of non violent and any recidivism. The AUC from the risk category was found to be greater at predicting general recidivism than the total score. The authors did surmise, however, that the homogeneity of the sample could have resulted in restricted scores and thus be affecting the predictive accuracy.
•Marshall et al. (2006) found moderate AUC values across three offending behaviours: recorded incidents of violence (.61), number of charges and convictions (.71) and assaults (.67).
•Testing the YLS/CMI on 254 justice involved youths of both genders found that the total score generated acceptable inter rater reliability at .77 (Scott, Brown and Skilling, 2019).
•Olver, Stockdale and Wong (2012) the YLS/CMI achieved moderate to high accuracy in predicting both youth and adult recidivism (i.e. general, non violent and violent recidivism) in a sample of male youths with AUC values ranging between 69. and .85.
•Latessa et al. (2016) found that inter rater reliability was acceptable for the Total Risk Score at 0.77; however, this fell below acceptable Kappa standards to 0.53 for the Overall Risk Level.
•A study of the YLS/CMI in Scotland found that the AUCs generated for general recidivism were 0.72 and 0.73 for males and females respectively. The AUCs for serious violence were 0.66 for males and 0.69 for females, suggesting that the tool is more accurate in predicting general recidivism (Vaswani, 2013).
•Vaswani and Merone (2012) examined 1138 YLS/CMI assessments in Scotland. The composite score achieved moderate to high AUC values in predicting ‘any’ (.73) and ‘serious violent’ recidivism (.68) in a sample of male
•Viljoen et al. (2009) the YLS/CMI composite score appeared to have moderate predictive accuracy for non sexual violent reoffending (.68), any violent reoffending (.61) and ‘any’ reoffending (.66)
•McGrath and Thompson (2012) in a one year follow up the the YLS/CMI obtained an AUC value of .65 for any re offence in a sample of Australian youth.
•Salekin (2008) found moderate ROC values of .66 and .64 for general and violent recidivism in a sample of male and female adolescents.
•Stockdale (2008) found large ROC values of .79 for general recidivism and .78 for violent recidivism.
b)International Research
•Chu et al. (2016) found total scores significantly predicted general recidivism for both male and female youth in a Singapore study.
•Vieira et al. (2009) reported that youths who had less than a third of their identified criminogenic needs met were eighteen times more likely to reoffend in a three year follow up period in comparison to youths for whom the majority of their needs were met.
Scottish youth. Results indicated that the tool was also a good predictor for the occurrence, speed and volume of reoffending for males and females. The AUCs for total scores were .73 and .72 for males and females respectively. For males and females in the risk/needs category the score were .68 and .69 respectively.
•Onifade, et al. (2008a) found that the YLS/CMI correctly classified 59% of individuals as either recidivists or non recidivists.
•After carrying out AUC analyses, Latessa et al. (2016) found that the YLS/CMI did not predict youth recidivism much better than chance for the overall sample and the sample divided up by gender.
•Olver, Stockdale and Wormith (2009) moderate correlation found between the YLS/CMI score and general recidivism (r = .32).
•In a sample of 254 youth, Scott, Brown and Skilling (2019) found that the YLS/CMI total risk score was .68 and the total strength score yielded smaller effects at .59.
RATED page updated: July 2019 © Risk Management Authority 2019
•A study into juveniles in a Spanish province using a translated version of the YLS/CMI found that gender played a significant role in affecting recidivism risk levels: females assessed by the YLS/CMI as having a low risk level had a higher risk of recidivism than boys (Jara, García Gomis and Villanueva, 2016).
b)International Research
Applicability: Females
a)UK Research
•Vaswani and Merone (2012) the composite score achieved moderate to high AUC values in predicting ‘any’ (.72) and ‘serious violent’ recidivism (.69) in a sample of female Scottish youth.
•Marshall et al. (2006) the YLS/CMI demonstrated moderate correlations with the number of charges and convictions (r = .32) and assaults (r =.40) in a sample of females.
•Bechtel, Lowenkamp and Latessa (2007) small correlations observed between the composite score and recidivism in the total female sample (r = .17); however, in community samples, no significant correlations were found.
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
•Olver, Stockdale and Wong (2012) observed moderate to high accuracy in predicting both youth and adult recidivism (i.e. general, non violent and violent recidivism) in a sample of female youth with AUC values ranging between 65. and .75.
•Anderson et al. (2016) claimed that predictive validity of the tool is affected by gender with AUCs of .623 and .565 for boys and girls respectively. In their study, it was found that only the family and personality subscales significantly predicted recidivism for girls compared to all the subscales for boys.
•Using a sample of 440 juveniles in Australia, McGrath, Thompson and Goodman Delahunty (2018) tested the predictive validity of the Australian adaption of the tool, YLS/CMI AA (Hoge and Andrews, 1995). Predictive accuracy differed by gender, with AUCs of .694 and .667 generated for males in relation to general and violence recidivism respectively. An AUC of .690 was yielded for general recidivism in females; whilst the AUC for violence recidivism was higher than males at .725.
•The YLS/CMI total risk score showed good predictive accuracy for males and females, with AUCs of .68 for both genders. The total strength score was found to yield a smaller effect for males at .60 compared to .69 for females (Scott, Brown and Skilling, 2019).
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
•A study found that Spaniards to indicated a higher level of reoffending than ethnic minorities in the follow up period. Protective factors, however, negated risk factors across different ethnic groups. To that end, when a youth possessed protective factors, their nationality no longer had an impact on their reoffending (Cuervo and Villanueva, 2018).
•After an examination of 334 YLS/CMI assessments, Perrault et al. (2017) found that race was not a significant factor in predict reoffending; although Black youth did score higher on the official juvenile history scale than White young people.
a)UK Research
•Olver, Stockdale and Wong (2012) observed moderate accuracy in predicting both youth and adult recidivism (i.e. general, non violent and violent recidivism) in a sample of Aboriginal youth with AUC values ranging between .62 and .67.
•A study by the National Council on Crime and Delinquency in the United States found the YLS/CMI did not perform as well for Black/African Americans and Hispanic/Latinos (Baird et al., 2013).
•Chu et al. (2012) in a sample of 165 male youths from Singapore, higher mean scores on the YLS/CMI were
Applicability: Ethnic Minorities
None available at present.
•Taylor (2018) carried out statistical analyses on 1679 youth looking at various social factors, including gender. It was discovered that females demonstrated significantly higher needs on the personality/behaviour and family circumstances/parenting subcomponents compared to males, indicating that strained and stressed family relationships are a significant area of risk and need for females offending
•Onifade, et al. (2008b) the tool achieved 60% accuracy in identifying recidivists and non recidivists in an African American sample of males. .
Validation History
Applicability: Mental Disorders
a)UK Research
•Onifade et al. (2009) found no significant differences in the predictive validity of the YLS/CMI between racial groups (Caucasian AUC = .66 versus African American AUC = .63).
•Bechtel, Lowenkamp and Latessa (2007) small correlations observed between the composite score and recidivism of non White individuals in community (r= .23) and institutionalised (r =.10)
b)International Research
RATED page updated: July 2019 © Risk Management Authority 2019
observed among individuals affiliated to gangs compared to those who were not affiliated to gangs.
•Rennie and Dolan (2010) found that recidivists attained higher scores than non recidivists on their past and current offending. The YLS/CMI generated AUCs of .60, .66 and .67 for violent, non violent and ‘any’ recidivism respectively in a sample of mentally disordered males.
•McGrath, Thompson and Goodman Delahunty (2018) utilised a sample of 440 juveniles in Australia to test differences within ethnic subgroups using the Australian adaption of the YLS/CMI. For general recidivism, AUCs were generated of .648, .684 and .716 for indigenous, non indigenous and ethnic groups respectively. There were similar findings for violence recidivism: indigenous, .623; non indigenous, .668; ethnic, .715.
•McLachlan et al. (2018) examined the predictive validity of the YLS/CMI to measure recidivism in 100 youth: half of the sample had ‘foetal alcohol spectrum disorder’ (FASD); the other half did not have FASD or prenatal
•Villanueva and colleagues (2019) administered a Spanish translation of the YLS/CMI to young Spanish individuals, 116 of which were Arab and 140 who were non Arab. With the inclusion of subtle cultural differences, AUCs of .73 and .76 were generated for the Arab and non Arab groups respectively. To that end, the YLS/CMI was able to predict the correct outcomes for 73.7% of Arab and 75.9% of non Arab minor individuals respectively.
•The YLS/CMI can aid assessors in identifying risk, responsivity and protective factors specific to the individual.
RATED page updated: July 2019 © Risk Management Authority 2019
•Campbell et al. (2014) explored the use of a reduced item YLS/CMI as a brief screener with positive results. In the course of their study, the YLS was confirmed to be gender neutral.
•A meta analysis conducted by Schwalbe (2008) on a number of youth risk assessment tools found small differences in effect sizes for the YLS/CMI between males and females. The author suggests that gender differences observed in individual studies, may evidence ‘…gender biases in juvenile
•The YLS/CMI seems to be neutral in terms of gender and race/ethnicity (Barnes et al., 2016).
•Baird et al. (2013) had a few criticisms of a couple of the items within the YLS/CMI. The ‘could make better use of time’ factor within the leisure/recreation domain is said to be a subjective item that is difficult to reliably score. Within the substance abuse item, there are options to check for ‘occasional drug use’ and ‘chronic drug use,’ which would appear to be mutually exclusive. In spite of this, both of these are to be selected when chronic drug use is checked. This is something which always happens in automated versions of the YLS/CMI; yet, it is not consistently applied when the scoring is carried out manually by assessors, something which leads to scoring errors. For instance, in Nebraska commitment cases, workers neglected to comply with this rule in 12.3% of cases.
•Several studies found that usage of the professional override function reduced the accuracy of the YLS/CMI to predict general recidivism, particularly in serious violence cases. For instance, the AUC for serious violence recidivism was .68 when using the YLS/CMI; this was then reduced to an AUC of .54 when the professional override was used (Vaswani and Merone, 2012). Based on this, it is advised that the ‘professional override’ function should be used with extreme caution (Schmidt et al., 2016; Vaswani ,2013).
•Information obtained from this tool can contribute to risk management strategies.
•The authors recommend that YLS/CMI measurements are updated every six months to capture the dynamic nature of youth development.
Contribution to Risk Practice
•Regression analyses found the best predictors of recidivism in the tool were the following risk factors: school or employment problems, criminal friends and personality behaviour (Cuervo and Villanueva, 2018).
Other Considerations
•Vaswani (2013) found that when the YLS/CMI was used with people aged over 18 is yielded no statistical significance; hence, it was not very accurate in predicting reoffending. It is, thus, recommended that the tool is not used with those aged 18 and over and that the adult version of the tool is instead used (Vaswani and Merone, 2012).
•The tool can contribute towards measuring progress or deterioration in factors related to the individual’s level of risk.
•The dynamic factors included in the YLS/CMI can act as targets for change.
alcohol exposure. Results showed the YLS/CMI was able to predict recidivism in youth with FASD; although this group was rated at higher risk across all risk ratings, suggesting a high level of risk and intervention need.
•Qualitative analysis by Burman and colleagues (2007) highlighted the limitations of the YLS/CMI in relation to its inability to discern the type and severity of offending behaviours and the lack of a separate ‘Risk of Harm’ Section.
•An online version is available through the distributors MHS.For more information on the YLS/CMI 2.0 http://www.mhs.com/product.aspx?gr=safandprod=ylsvisit: cmi2andid=overview
justice decision making and case processing rather than for the ineffectiveness of risk assessment with female offenders…’ (Schwalbe, 2008: 1367)
•The revised version (YLS/CMI 2.0; Hoge & Andrews, 2010) has been published and contains important new developments which include but are not limited to the following: (1) new recommended cut off scores for different risk/need levels; (2) expanded age range (12 18); (3) inclusion of items addressing gender informed responsivity factors like pregnancy/motherhood.
RATED page updated: July 2019 © Risk Management Authority 2019
•The scoring of the ‘Total Risk/Need Score and the eight subcomponents of Part I (Assessment of Risks and Needs) remains unchanged from the YLS/CMI (Hoge and Andrews, 2010: 3).
Name of Tool AssetPlus
Category
•Factors included can contribute towards measuring progress or deterioration in relation to the individual’s level of risk. This includes a measurement of whether a young person is susceptible to exploitation or manipulation.
Assessors are required to complete a detailed induction programme on AssetPlus. Refresher training every three to six months is highly likely, particularly in the areas of Risk of Serious Harm, Safety and Wellbeing and identifying desistance factors.
RATED page updated: July 2019 © Risk Management Authority 2019
Tool Development
•"The AssetPlus framework will provide a single assessment and plan for a young person, which will be dynamic and iterative in nature, making it easier to update assessments on an ongoing basis and therefore always presenting the latest information" (Baker, 2014).
10 18
Author / Publisher Baker and Youth Justice Board
•Designed to be used by Youth Offending Teams (YOT) in England and Wales. ‘Youth Offending Team’ (YOT) transfer questions are included in AssetPlus to support case transfers between teams.
•AssetPlus is a structured assessment tool for use with young individuals who come into contact with the criminal justice system. It is designed to contribute to assessment and intervention planning for youths in both custodial and community settings. It has superseded the ‘Asset’ instrument.
Description
Assessor Qualifications
Age Appropriateness
Year 2014
•The instrument has scope to record whether a young person is susceptible to being manipulated or exploited.
•The instrument examines the young person’s offence history and identifies a multitude of factors or circumstances which may have contributed to the behaviour. This includes questions about the inappropriate use of technology, health questions like whether the young person could be pregnant and a prompt about offence paralleling behaviours. It will also highlight any particular needs or difficulties for intervention as well as changes in risk and need over time.
Youth Assessment: General Risk (Awaiting Validation)
•The AssetPlus builds upon the original Asset instrument. It uses the YOGRS as a static predictor to provide an indication of likely re offending This should help promote greater alignment with the National Probation Service’s assessment framework (Baker, 2014).
RATED page updated: July 2019 © Risk Management Authority 2019
•Asset has been used in a number of local authorities since 2001. In Scotland, the processing of police charges through the Children’s Hearing System means that only the dynamic component of the Asset tool is used in everyday practice. To clarify, it is not possible to use the static component of the instrument, whereby involvement in previous offending is scored in relation to convictions, something which tends to be negated by the involvement of the Children’s Hearing System in Scotland. To negate this, changes were made to the terminology in the ‘Asset’ when used in Scotland (Fearn 2014).
•The instrument can aid assessors in identifying risk and responsivity factors specific to the individual (e.g. ‘motivation to change’). Some of the factors included in the assessment can act as targets for change. The AssetPlus can also help assessors identify protective factors and strengths.
•Participation of the young person in identifying and responding to their behaviours is part of the structure of the instrument.
•Empirical research on the AssetPlus is pending. Currently, there has just been research on the Asset instrument (Baker et al., 2003; Baker et al., 2005; Burman et al., 2007).
General Notes
Author / Publisher Levene and Colleagues
•Items are rated on a 3 point scale from 2 for presence, 1 for some but not complete and 0 for the lack of presence. There is also a clinical risk column for the assessor to red flag any factors that are particularly concerning (Augimeri et al., 2010a).
•There are three sections within the tool: 1) family, looking at familial support and stressors; 2) child, looking at risk factors associated with the individual; 3) responsivity in terms of the ability and willingness of the individual and their family members to engage in interventions (Augimeri et al., 2001; 2019).
Year 2001
•It is similar to the boys’ equivalent EARL 20B, with the inclusion of some gender responsive items such as caregiver daughter interaction.
RATED page updated: July 2019 © Risk Management Authority 2019
•Girls had significantly higher composite scores than boys on the 19 common items shared in both the EARL 20B and EARL 21G assessments particularly on the ‘Family’ and ‘Child’ items (Augimeri et al., 2010).
Age Appropriateness
For girls ages 12 and under Assessor Qualifications
Description
•Developed from adult assessment tools and juvenile offending screening assessments (Augimeri et al., 2005).
•de Ruiter and Augimeri (2012) the EARL 21G achieved moderate to strong predictive accuracy in relation to teacher reported delinquency and the composite scores (AUC=.68) and final risk judgement (AUC = .71). No significant association was found, however, between EARL 21G scores and recidivism in official police records.
•The EARL 21G is a 21 item structured clinical risk assessment tool to be used with females aged 6 12 years. It is designed to assess the risk level of future anti social behaviour in order to inform treatment planning.
Category Youth Assessment: General Risk (Awaiting Validation)
Tool Development
•The goal of the EARL 21G is to help clinicians determine effective clinical risk management plan that may negate risk and prevent high risk children entering the juvenile or adult justice systems (Augimeri et al., 2010).
EARL 21G should be used by clinicians and other professionals with experience of working with high risk children.
Name of Tool Early Assessment Risk List for Girls (EARL 21G)
RATED page updated: July 2019 © Risk Management Authority 2019
General Notes
•The EARL 21G had been used in Edinburgh and is currently used in Glasgow as part of the SNAP® (Stop Now And Plan)pilot programme, a multi modal, gender specific, evidence based intervention for young children aged 6 11 with conduct problems and their families Augimeri et al., 2017; 2018). A presentation by Augimeri, Walsh and Donato (2016) explored the criminal outcomes for participants of SNAP®. The first wave informed the cost benefit analysis of SNAP® and the second wave informed the trajectory of SNAP® children and the predictive validity of the EARL tools.
•Augimeri et al. (2010b) found that only the item ‘Antisocial Values and Conduct’ predicted criminal outcomes for a sample of 380 girls.
•The EARL Pre Checklist (EARL PC: Augimeri et al., 2010a) has been recently developed as an abbreviated version of the full EARL assessments (see the EARL 20B entry in the Responsivity category for more information).
•Yuile (2007) the EARL 21G attained low to high kappa values ranging from .34 to .88 with an averaged item level agreement on individual items of .55. Reliability scores for four of the items were low (i.e. .36 to .40).
•Koegl (2011) the composite score for the EARL 21G attained moderate accuracy in predicting any offence (AUC = .65).
•Fewer studies examining the predictive accuracy of the final judgement ratings.
•The majority of the current validation literature has been conducted by the authors of the EARL 21G.
•Levene et al. (2001) moderate to high ICC values obtained for the EARL 21G composite scores (ICC = .67 [single measure] and .86 [average measure]).
•Other studies have used factor analysis methodology to validate the underlying constructs relating to the tool (e.g. confirmatory factor analysis used to test data based on existing theory or analytic research).
•No validation evidence for UK samples at present.
•Augimeri et al (2010a) using cox regression analysis, to analyse the relationship between variables, it was found that scores on the ‘Antisocial Values and Conduct’ item were significantly related to an increased probability to engage in future criminal offences. This was more so than the composite scores containing a combined score of multiple variables
•The EARL 21G was used in the ‘Interventions for Vulnerable Youth’ Project funded by the Scottish Government.
•Koegl et al. (under review) found that overall, manifesting antisocial behaviour was the strongest predictor for future criminal convictions (OR = 6.00), and poor coping Ability (C12) was associated with more than a fourfold increase in the odds of committing an offence.
•The tool has been translated into translated into six languages other than English (Swedish, Finnish, Norwegian, French, Dutch, and Japanese) and is used in various countries to assess the risk of future antisocial behaviours in young children.
12 18 years
•The YLS/CMI consists of 42 items relating to the ‘Central Eight’ risk and need domains. These are the social circumstances and personality items considered to be reflective of recidivism (Andrews, Bonta and Wormith, 2004). The YLS/CMI SV is an abbreviated version of this tool, corresponding to the eight risk/needs domains of the YLS/CMI (Hoge & Andrews, 2009).
•A study applying the YLS/CMI SV and the YLS/CMI tools to Singaporean youths gave a fair inter rater reliability rating of .51 for the YLS/CMI SV compared to a good one of .63 for the YLS/CMI The authors note that this may be a result of the small sample size for testing inter rater reliability (Chu et al., 2014).
General Notes
Category Youth Assessment: General Risk (Awaiting Validation)
•As an abbreviated version of the YLS/CMI, the YLS/CMI SV consists of 8 items pertaining to the risk/need domains of the tool: history of conduct disorder, current school or employment problems, some criminal friends, alcohol/drug problems, leisure/recreation, personality/behaviour, family circumstances/parenting and attitudes/orientation. Items are scored as 1 for the presence of a risk factor and 0 for the absence of it (Chu et al., 2014; Cuervo and Villanueva, 2017).
•The AUCs generated from the YLS/CMI SV when applied to 3264 youth in Singapore were .64, .63 and .61 for predicting general, non violent and violent recidivism respectively (Chu et al., 2014).
Age Appropriateness
•Cuervo and Villanueva’s (2017) study examining the ‘reduced version’ of the YLS/CMI on a sample of Spanish youths found that the AUC score for predicting general recidivism was .775.
•There have been a handful of international studies examining the reliability and predictive power of the YLS/CMI SV.
Name of Tool Youth Level of Service/Case Management Inventory 2.0 Screening Version (YLS/CMI SV 2.0)
Year 2009 Description
Tool Development
Author / Publisher Hoge and Andrews
•The screening version of the YLS/CMI may be used as a preliminary assessment for identifying young people at risk of general recidivism.
Assessors should possess training and experience in youth assessment
Assessor Qualifications
RATED page updated: July 2019 © Risk Management Authority 2019
•The YLS/CMI has been adopted as the primary risk assessment measure for youth in Singapore. Several justice agencies are contemplating whether to also use the YLS/CMI SV to screen for risk of general recidivism (Chu et al., 2014).
•For young person is assessed as being at moderate or high risk of recidivism using the YLS/CMI SV, a fuller assessment using the YLS/CMI would be warranted (Chu et al., 2014).
RATED page updated: July 2019 © Risk Management Authority 2019
RATED page updated: July 2019 © Risk Management Authority 2019
Baker,AssetPlusK.,
Yates, P.M. (2005) Pathways to the Treatment of Sexual Offenders: Rethinking Intervention Forum, Summer Beaverton OR: Association for the Treatment of Sexual Abusers, 1 9. [Not Augimeri,EARLaccessible]20BL.
Baker, K., Jones, S. and Roberts, C. (2005). Further Development of ASSET. Youth Justice Board for England and Wales. London: Youth Justice Board. [Not accessible]
Augimeri, L. Koegl, C., Levene, K., Webster, C. (2005). Early Assessment Risk Lists for Boys and Girls. In Thomas Grisso, Gina Vincent, Daniel Seagrave. (2005) Mental health Screening and Assessment in Juvenile Justice. New York, New York: The Guilford Press, 295 310. Access Here.
K., Koegl, C., Webster, C. and Levene, K. (2001). The Early Assessment of Risk List for Boys (EARL 20B) (Version 2). Toronto: Earlscourt Child and Family Centre. [Not accessible]
VALIDATED TOOLS
Burman, M., Armstrong, S., Batchelor, S., McNeill, F. and Nicholson, J. (2007). Research and Practice in Risk Assessment and Risk Management of Children and Young People Engaging in Offending Behaviours: A Literature Review. Glasgow, UK: The Scottish Centre for Crime and Justice Research. Access Here.
Augimeri, L. K., Pepler, D., Walsh, M. M., Jiang, D. and Dassinger, C. (2010). Aggressive and antisocial young children: Risk prediction, assessment and clinical risk management. Program Evaluation Report submitted to The Provincial Centre of Excellence for Child and Youth Mental Health at CHEO (Grant: # RG 976). Toronto, Ontario: Centre for Children Committing Offences and Program Development. Access Here
Fearn, G. (2014) Youth Crime: Am Investigation into the Effectiveness of General Reoffending Risk Assessment Tools. Unpublished ForenPsyD thesis. Birmingham, UK: Centre for Forensic Criminological Psychology, University of Birmingham. Access Here
Jones, S., Roberts, C. and Merrington, S. (2003). The evaluation of the validity and reliability of the Youth Justice Board’s assessment for young offenders. Findings for the first two years of the use of ASSET. London: Youth Justice Board. [Not accessible]
Augimeri, L. K., Pepler, D., Walsh, M. and Kivlenieks, M. (2017) ‘Addressing Children’s Disruptive Behavior Problems: A Thirty Year Journey with SNAP (Stop Now And Plan).’ In Sturmey, P. (Ed.), Handbook of Violence and Aggression, Volume 2: Assessment, Prevention, and Treatment of Individuals. U.S.: Wiley Blackwell. Access Here
Baker, K. (2014) AssetPlus Rationale. London: Youth Justice Board. Access Here
Augimeri, L. K., Walsh, M., Jiang, D., Koegl, C. J. and Logue, C. (2010) Early Assessment Risk List PreChecklist: EARL PC (Pilot Checklist). Toronto, ON: Child Development Institute. [Not accessible]
Augimeri, L. K., Walsh, M. M., Liddon, A. D. and Dassinger, C. R. (2011) ‘From risk identification to risk management: a comprehensive strategy for young children engaging in antisocial behavior. In Springer, D. W. and Roberts, A. (eds). Juvenile Justice and Delinquency. Sudbury, Massachusetts: Jones and Barlett Publishers, 117 140. Access Here
Koegl, C. J. (2011) ‘High risk antisocial children: predicting future criminal and health outcomes. Unpublished doctoral thesis. Cambridge, UK: University of Cambridge. [Not accessible]
Walsh, M., Enebrink, P., Jiang, D., Blackman, A., & Kanter, D. (2019). Gender specific childhood risk assessment tools: Early Assessment Risk Lists for Boys (EARL 20B) and Girls (EARL 21G). In Otto, R. K. and Douglas, K. S. (Eds.), Handbook of violence risk assessment, 2nd edition, 43 62. Oxford, UK: Routledge, Taylor & Francis. [Not accessible]
Levene, K. S., Augimeri, L. K., Pepler, D., Walsh, M., Webster, C. D. and Koegl, C. J. (2001). Early Assessment Risk List for Girls Version 1, Consultation Edition (EARL 21 G). Toronto: Earlscourt Child and Family Centre. [Not accessible]
de Ruiter, C., van Domburgh, L. and Augimeri, L. (2012) ‘Risk assessment of antisocial behavior in children under 12: The Early Assessment Risk Lists.’ In Lodewijks, H. P. B. and van Domburgh, L.(eds.) Instruments for risk assessment of harm by and in children and youth. Amsterdam: Pearson Assessment. [Not accessible]
De Ruiter, C. and Augimeri, L. K. (2012) ‘Making delinquency prevention work with children and adolescents: From risk assessment to effective interventions.’ In Logan, C. and Johnstone, L. (eds.) Managing Clinical Risk: A Guide to Effective Practice, 199 223. London: Routledge. Access Here
Augimeri, L. K., Walsh, M., Donato, A., Blackman, A. and Piquero, A. R. (2018) ‘SNAP (Stop Now And Plan): Helping children improve their self control and externalizing behavior problems.’, Journal of Criminal Justice: Special Issue Advances in Research on Self Control, 56, 43 49. Access Augimeri,Here.L.K.,
De Ruiter, C. and van Domburgh, L. (2016). Predictive Validity of the Early Assessment Risk List for Boys (EARL 20B) after First Police Contact: Differences between Western and Non Western Boys in The Netherlands. Manuscript in preparation.
Koegl, C.J., Farrington, D. and Augimeri, L. (submitted). Predicting Future Criminal Convictions in Children Under Age 12 using The Early Assessment Risk Lists. Journal of Developmental and Life Course Criminology. Journal of Developmental and Life Course Criminology, Special issue: Developmental and life course approaches to crime prevention. [Not accessible]
Hrynkiw Augimeri, L. K. (2005). Aggressive and Antisocial Young Children: Risk Assessment and Management Utilising the Early Assessment Risk List for Boys (EARL 20B). Unpublished PhD Dissertation. Toronto, Canada: University of Toronto. [Not accessible]
Webster, C., Douglas, K. S., Eaves, D. and Hart, D. (1997). HCR 20 assessing risk for violence: Version II. Burnaby, British Columbia, Canada: Simon Fraser University. [Not accessible]
Enebrink, P., Långström, N., Hultén, A. and Gumpert, C. H. (2006b). Swedish Validation of the EARL 20B: A Decision Aid for use with Children Presenting with Conduct Disordered Behaviour. Nordic Journal of Psychiatry, 60, 438 446. Access Here.
RATED page updated: July 2019 © Risk Management Authority 2019
Enebrink, P., Långström, N., and Gumpert, C.H. (2006a). Predicting aggressive and disruptive behaviour in referred 6 12 year old boys: Prospective validation of the EARL 20B risk/needs checklist. Assessment, 13(3), 356 367. Access Here
Andrews, D. A., Bonta, J. and Wormith, S. (2011) ‘The Risk Need Responsivity Model.’ Criminal Justice and Behavior 38 (7), 735 755. Access Here
YLS/CMI Anderson,2.0V. R., Davidson II, W. S., Barnes, A. R. Campbell, C. A., Peterson, J. L. and Onifade, E. (2016) ‘The differential predictive validity of the Youth Level of Service Case Management Inventory: the role of gender.’ Psychology, Crime and Law 22(7), 666 677. Access Here
Washington State Juvenile Court Assessment Manual, Version 2.1 (Report No. 04 03 1203). Olympia, WA: Washington State Institute for Public Policy. [Not accessible]
Scott, T., Brown, S. I. and Skilling, T. A. (2019) ‘Predictive and Convergent Validity of the Youth Assessment and Screening Instrument in a Sample of Male and Female Justice Involved Youth.’ Criminal Justice and Behavior 46(6), 811 831. Access Here
Healy, Kristen Johnson, Andrea Bogie, Erin Wicke Dankert and Chris Scharenbroch. (2013) A comparison of risk assessment instruments in juvenile justice. Washington, D. C.: National Council on Crime and Delinquency, U.S. Department of Justice. Access Baronski,HereR.(2004)
Jones, N. J., Brown, S. L., Wanamaker, K. A., and Greiner, L. E. (2014) ‘A Quantitative Exploration of Gendered Pathways to Crime in a Sample of Male and Female Juvenile Offenders.’ Feminist Criminology 9(2), 113 136. Access Here
RATED page updated: July 2019 © Risk Management Authority 2019
Walsh, M., Yuile, A., Jiang, D., Augimeri, L.K., Pepler, D. (2007). Early Assessment Risk List for Girls (EARL 21G): Predicting Antisocial Behaviors and Clinical Implications. Manuscript in Baird,YASIpreparation.ChrisTheresa
Jones, N. J., Brown, S. L., Robinson, D., and Frey, D. (2015) ‘Incorporating Strengths Into Quantitative Assessments of Criminal Risk for Adult Offenders: The Service Planning Instrument.’ Criminal Justice and Behavior 42(3), 321 338. Access Here.
Jones, N. J. (2013). Expanding “What Works”: The role of strengths in risk assessment. Poster presented at the annual convention of the American Psychology and Law Society, Portland, Oregon. [Not accessible]
Robinson, D. and Jones, N. (2017) The validity of youth assessment and screening instrument for justice involved youth in Milwaukee county. Ottawa, Canada: Orbis Partners Inc. [Not accessible]
Jones, N. J. Shelley L. Brown, David Robinson and Deanna Frey. (2016) ‘Validity of the Youth Assessment and Screening Instrument: A Juvenile Justice tool incorporating risk, needs and strengths.’ Law Human Behavior 40(2), 182 194. Access Here
Skeem, J., Kennealy, P., Hernandez, I., Tatar, J., Clark, S., Tartar, J., and Keith, F. (2013). CA YASI predictive utility: How well do scores and classifications predict youths' infractions and arrests? Report prepared for the Division of Juvenile Justice, California Department of Corrections and Rehabilitation (CDCR). [Not accessible]
Andrews, D. A. and Bonta, J. (1994). The psychology of criminal conduct. Cincinnati, OH, US: Anderson Publishing Co. Access Here
C. M., Daffern, M., Thomas, S. and Lim, J. Y. (2012) ‘Violence risk and gang affiliation in youth offenders: a recidivism study.’ Psychology, Crime & Law 18(3), 299 315. Access Here
Lux, Jennifer, Chouhy, Cecilia and Long, Joshua. (2016) Examining the Validity and Reliability of the Youth Level of Service/Case Management Inventory 2.0 for the South
Gomis, A. and Villanueva, L. (2016) ‘Impact of type of intervention on youth reoffending: are gender and risk level involved?’ Psychiatry, Psychology, Law 23(2), 215 223. Access Latessa,Here.Edward,
Campbell, C., Onifade, E., Barnes, A., Peterson, J., Anderson, V. Davidson, W. and Gordon, D. (2014) ‘Screening Offenders: The Exploration of a Youth Level of Service/Case Management Inventory (YLS/CMI) Brief Screener.’ Journal of Offender Rehabilitation 53(1), 19 34. Access
Barnes, A. R., Campbell, N. A., Anderson, V. R., Campbell, C. A., Onifade, E. and Davidson, W. S. (2016) ‘Validity of initial, exit and dynamic juvenile risk assessment: an examination across gender and race/ethnicity.’ Journal of Offender Rehabilitation 55(1), 21 38. Access Here.
Chu,Here
RATED page updated: July 2019 © Risk Management Authority 2019
Bechtel, K., Lowenkamp, C. T. and Latessa, E. (2007) ‘Assessing the Risk of Re Offending for Juvenile Offenders Using the Youth Level of Service/Case Management Inventory.’ Journal of Offender Rehabilitation 45(3 4), 85 108. Access Here
Hoge, R. D. and Andrews, D. A. (1995) Australian adaptation of the Youth Level of Service/Case Management Inventory. North Tonawanda, NY: Multi Health Systems. Access Here.
Baird, C., Healy, T., Johnson, K., Bogie, A., Wicke Dankert, E. and Scharenbroch, C. (2013) A comparison of risk assessment instruments in juvenile justice. Washington, D. C.: National Council on Crime and Delinquency, U.S. Department of Justice. Access Here
Burman, M., Armstrong, S., Batchelor, S., McNeill, F. and Nicholson, J. (2007). Research and Practice in Risk Assessment and Risk Management of Children and Young People Engaging in Offending Behaviours: A Literature Review. Glasgow, UK: The Scottish Centre for Crime and Justice Research. Access Here
Chu, C. M., Goh, M. L. and Chong, D. (2016) ‘The Predictive Validity of Savry Ratings for Assessing Youth Offenders in Singapore: A Comparison With YLS/CMI Ratings.’ Criminal Justice and Behavior 43(6), 793 810. Access Here.
Hoge, R. D. and Andrews, D. A. (2002). The Youth Level of Service/Case Management Inventory manual and scoring key. Toronto, ON: Multi Health Systems. [Not accessible]
Hoge R. D. and Andrews D. A. (2011). Youth Level of Service/Case Management Inventory 2.0 (YLS/CMI 2.0): User’s manual. Toronto, Ontario, Canada: Multi Health Systems. [Not Jara,accessible]P.,Garcia
Cuervo, K. and Villanueva, L. (2018) ‘Prediction of Recidivism With the Youth Level of Service/Case Management Inventory (Reduced Version) in a Sample of Young Spanish Offenders.’ International Journal of Offender Therapy and Comparative Criminology 62(11), 3562 3580. Access Here.
Marshall. J., Egan. V., English., E. and Jones. R. (2006) ‘Relative Validity of Psychopathy Versus Risk/Needs Based Assessments in the Prediction of Adolescent Offending Behaviour.’ Legal and Criminological Psychology 11(2), 197 210. Access Here
Perrault, R. T., Vincent G. M. and L. S. Guy. (2017) ‘Are risk assessments racially biased? Field study of the SAVRY and YLS/CMI in probation.’ Psychological Assessment 29 (6), 664 678. Access Here.
Pusch, N. and Holtfreter, K. (2018) ‘Gender and Risk Assessment in Juvenile Offenders: A Meta Analysis.’ Criminal Justice and Behavior 45(1), 56 81. Access Here.
Olver, M. E., Stockdale, K. C., and Stephen Wormith, J. (2009) ‘Risk Assessment With Young Offenders: A Meta Analysis of Three Assessment Measures.’ Criminal Justice and Behavior 36(4), 329 353. Access Here.
Dakota Department of Corrections Juvenile Community Corrections. Cincinnati, Ohio: University of Cincinnati Corrections Institute. Access Here.
Olver, M. E., Stockdale, K. C., and Wong, S. C. P. (2012) ‘Short and long term prediction of recidivism using the youth level of service/case management inventory in a sample of serious young offenders.’ Law and Human Behavior 36(4), 331 344. Access Here
Onifade, E., Davidson, W. and Campbell, C. (2009) ‘Risk Assessment: The Predictive Validity of the Youth Level of Service Case Management Inventory with African Americans and Girls.’ Journal of Ethnicity in Criminal Justice 7(3), 205 221. Access Here
McGrath, Andrew J., Thompson, Anthony P. and Goodman Delahunty, Jane. (2018) ‘Differentiating Predictive Validity and Practical Utility for the Australian Adaptation of the Youth Level of Service/Case Management Inventory.’ Criminal Justice and Behaviour 45(6), 820 834. Access McLachlan,Here.Kaitlyn,
Davidson, W., Campbell, C., Turke, G., Malinowski, J. and Turner, K. (2008b) ‘Predicting Recidivism in Probationers With the Youth Level of Service Case Management Inventory (YLS/CMI).’ Criminal Justice and Behavior 35(4), 474 483. Access Here.
RATED page updated: July 2019 © Risk Management Authority 2019
McGrath, A. and Thompson, A. P. (2012) ‘The Relative Predictive Validity of the Static and Dynamic Domain Scores in Risk Need Assessment of Juvenile Offenders. ‘Criminal Justice and Behavior 39(3), 250 263. Access Here.
Onifade, E., Davidson, W., Livsey, S., Turke, G., Horton, C., Malinowski, J., Atkinsin, D., Wimberly, D.(2008a) ‘Risk assessment: Identifying patterns of risk in young offenders with the Youth Level of Service/Case Management Inventory.’ Journal of Criminal Justice 36(2), 165 173. Access Onifade,HereE.,
Rennie, C. and Dolan, M. (2010) ‘Predictive validity of the Youth Level Of Service/Case Management Inventory in custody sample in England.’ Journal of Forensic Psychiatry and Psychology 21(3), 407 25. Access Here
Gray, Andrew L., Roesch, Ronald, Douglas, Kevin S. and Viljoen, Jodi L. (2018) ‘An evaluation of the predictive validity of the SAVRY and the YLS/CMI in justice involved youth with fetal alcohol spectrum disorder.’ Psychological Assessment 30(12), 1640 1651. Access Here
Viljoen, J. L., Elkovitch, N., Scalora, M. J., & Ullman, D. (2009) ‘Assessment of reoffense risk in adolescents who have committed sexual offenses: Predictive validity of the ERASOR, PCL:YV, YLS/CMI, and Static 99.’ Criminal Justice and Behavior 36(10), 981 1000. Access Here.
Salekin, R. T. (2008) ‘Psychopathy and recidivism from mid adolescence to young adulthood: Cumulating legal problems and limiting life opportunities.’ Journal of Abnormal Psychology 117(2), 386 395. Access Here
Vaswani, N. and Merone, L. (2012) ‘Are there risks with risk assessment? A study of the predictive accuracy of the youth level of service case management inventory with young offenders in Scoltand.’ The British Journal of Social Work 44(8), 2163 2187. Access Here.
Schmidt, F., McKinnon, L., Chattha, H. K., and Brownlee, K. (2006) ‘Concurrent and predictive validity of the Psychopathy Checklist: Youth version across gender and ethnicity.’ Psychological Assessment 18(4), 393 401. Access Here.
Schmidt, F., Sinclair, S. M. and Thomasdottir, S. (2016) ‘Predictive validity of the youth level of service/case management inventory with youth who have committed sexual and non sexual offences: the utility of professional override.’ Justice and Behavior 43(3), 413 430. Access Here.
Vaswani, N. (2013) The use of the YLS/CMI in Scotland. Glasgow, UK: Centre for Youth and Criminal Justice. No 01 (November). Access Here
Vieira T. A., Skilling T. A., Peterson Badali M. (2009) ‘Matching court ordered services with treatment needs: Predicting treatment success with young offenders.’ Criminal Justice and Behavior 36, 385 401. Access Here
Welsh, J. L., Schmidt, F., McKinnon, L., K., H. and Meyers, J. R. (2008) ‘A Comparative Study of Adolescent Risk Assessment Instruments: Predictive and Incremental Validity.’ Assessment 15(1), 104 115. Access Here
Schwalbe, C. S. (2008) ‘A Meta Analysis of Juvenile Justice Risk Assessment Instruments: Predictive Validity by Gender.’ Criminal Justice and Behavior 35(11), 1367 1381. Access Here.
Scott, T., Brown, S. I. and Skilling, T. A. (2019) ‘Predictive and Convergent Validity of the Youth Assessment and Screening Instrument in a Sample of Male and Female Justice Involved Youth.’ Criminal Justice and Behavior 46(6), 811 831. Access Here
Villanueva, L., Gomis Pomares, A. and Adrian, J. E. (2019) ‘Predictive Validity of the YLS/CMI in a Sample of Spanish Young Offenders of Arab Descent.’ International Journal of Offender Therapy and Comparative Criminology, 1 17. Access Here
RATED page updated: July 2019 © Risk Management Authority 2019
Stockdale, K. C. (2008) The Validity and reliability of the Violence Risk Scale Youth Version (VRs YV). Unpublished Doctoral Dissertation). Saskatoon, Canada: University of Saskatchewan Access Here.
Taylor, Alan. (2018) Risk Profiles and the YLS/CMI: Examining Mean Differences in Gender, Race, Hispanic Origin, Child Service History, Gangs and Offence Type Master of Social Work Degree, University of Manitoba: Winnipeg. Access Here
Augimeri, L. K., Koegl, C. J., Levene, K. S. and Webster, C. D. (2005) ‘Early Assessment Risk Lists for Boys and Girls.’ In Grisso, T., Vincent, G. and Seagrave, D. (eds.). Mental health screening and assessment in juvenile justice. New York, NY: Guilford Press, 295 310. Access Here.
Augimeri, L. K., Pepler, D., Walsh, M. and Kivlenieks, M. (2017) ‘Addressing Children’s Disruptive Behavior Problems: A Thirty Year Journey with SNAP (Stop Now And Plan).’ In Sturmey, P. (Ed.), Handbook of Violence and Aggression, Volume 2: Assessment, Prevention, and Treatment of Individuals. U.S.: Wiley Blackwell. Access Here
Augimeri, L. K., Walsh, M., Donato, A., Blackman, A. and Piquero, A. R. (2018) ‘SNAP (Stop Now And Plan): Helping children improve their self control and externalizing behavior problems.’, Journal of Criminal Justice: Special Issue Advances in Research on Self Control, 56, 43 49. Access Here.
Augimeri, L., Walsh, M. and Donato, A. (2016) ‘SNAP® (Stop Now and Plan) and Future Criminal Outcomes: A Case for Intervention During the Middle Years.’ International Association of Forensic Mental Health Services, June 21st 23rd. New York: Fordham University. Access Here
Koegl, C. J. (2011) High risk antisocial children. Predicting future criminal and health outcomes. Doctoral dissertation. Cambridge, UK: Institute of Criminology, University of Cambridge. [Not accessible]
RATED page updated: July 2019 © Risk Management Authority 2019
EARL 21G
Augimeri, L. K., Pepler, D., Walsh, M. M., Jiang, D. and Dassinger, C. (2010b). Aggressive and antisocial young children: Risk prediction, assessment and clinical risk management. Program Evaluation Report submitted to The Provincial Centre of Excellence for Child and Youth Mental Health at CHEO (Grant: # RG 976). Toronto, Ontario: Centre for Children Committing Offences and Program Development. Access Here.
Augimeri, L. K., Enebrink, P., Walsh, M. and Jiang, D. (2010a) ‘Gender Specific Childhood Risk Assessment Tools: Early Assessment Risk Lists for Boys (EARL 20B) and Girls (EARL 21G).’ In Otto, R. K. and Douglas, K. S. (eds.) Handbook of Violence Risk Assessment. London: Routledge, 43 62. Access Here.
de Ruiter, C. and Augimeri, L. K. (2012) ‘Making delinquency prevention work with children and adolescents: From risk assessment to effective interventions.’ In C. Logan and L. Johnstone (Eds.), Managing clinical risk: A guide to effective practice. London, UK: Routledge, 199 223. Access Here.
TOOLS AWAITING VALIDATION
Yates, P.M. (2005) Pathways to the Treatment of Sexual Offenders: Rethinking Intervention Forum, Summer. Beaverton OR: Association for the Treatment of Sexual Abusers, 1 9. [Not accessible]
Augimeri, L., Walsh, M., and Donato, A. (2016) SNAP (Stop Now And Plan) and future criminal outcome: A case for intervention during the middle years Part II. Symposium paper presented at the International Association of Forensic Mental Health Services Conference, New York, NY. Access Here.
Cuervo, K. and Villanueva, L. (2018) ‘Prediction of Recidivism With the Youth Level of Service/Case Management Inventory (Reduced Version) in a Sample of Young Spanish Offenders.’ International Journal of Offender Therapy and Comparative Criminology 62(11), 3562 3580. Access Here.
YLS/CMI SV 2.0 Chu, C. M., Yu, H., Lee, Y. and Zeng, G. (2014) ‘The utility of the YLS/CMI SV for Assessing Youth Offenders in Singapore.’ Criminal Justice and Behavior 41(12), 1437 1457. Access Here
RATED page updated: July 2019 © Risk Management Authority 2019
Levene, K. S., Augimeri, L. K., Pepler, D. J., Walsh, M. M., Webster, C. D. and Koegl, C. J. (2001) Early assessment risk list for girls, EARL 21G (Version 1). Toronto, ON: Earlscourt Child and Family Centre. [Not accessible]
Yuile, A. (2007) Developmental pathways of aggressive girls: A gender sensitive approach to risk assessment, intervention, and follow up. Doctoral dissertation. Toronto, Canada: York University. [Not accessible]
RATED page updated: July 2019 © Risk Management Authority 2019
Category Youth Assessment: Violence Risk (Validated)
•Also examined on the SAVRY are protective factors like prosocial involvement, strong social support, attachments and bonds, positive attitudes towards intervention and authority, strong commitment to school and resilient personality traits.
Assessors should possess training and experience in youth assessment, expertise in child/adolescent development and conducting risk assessments (Borum et al., 2010).
Description
•The items are clustered under three risk domains: (1) Historical Risk Factors, looking at history of violence, self harm and suicide attempts, and exposure to violence within the home; (2) Social/Contextual Risk Factors, focusing on peer delinquency and rejection, stress and poor coping skills, poor parental management, lack of personal support and community disorganisation; (3) Individual/Clinical Factors, examining negative attitudes, risk taking/impulsivity, substance use difficulties, anger management, lack of personal and social support.
12 18
•Provides a systematic approach to risk assessment which may assist in highlighting risk factors to be addressed in risk formulation and risk management planning.
•This tool considers dynamic variables as well as static ones. This allows for the assessment of change in risk level (i.e. progress in treatment) and also informs intervention needs and targets (Yates, 2005).
Name of Tool Structured Assessment of Violence Risk in Youth (SAVRY)
Age Appropriateness
•Interviews are carried out with the student and their family members as part of the assessment. Data from the mental health providers and physicians involved are also used.
•The SAVRY contains six additional protective factors as a separate set of items to risk factors. These are considered positive items notable for their presence (as opposed to negative protective factors significant for their absence) (Borum et al., 2006).
Assessor Qualifications
•The SAVRY is a 24 item structured assessment of violence risk in adolescents
•Designed for use with individuals aged between 12 and 18.
Strengths
Year 2006
•The SAVRY is not designed to be a formal test or scale to ‘quantify risk’; there are no assigned numerical values nor are there any specified cut off scores. The purpose of SAVRY is to provide operational definitions of risk factors for examiners to apply (Borum et al., 2010).
Author / Publisher Borum, Bartel and Forth
•Majority of items can be coded using file information.
•McGowan et al. (2011) found ICC for both raters in the study (.81).
RATED page updated: July 2019 © Risk Management Authority 2019
Empirical Grounding
a)UK Research
•Dolan and Rennie (2008) the SAVRY was found to have excellent inter rater reliability in relation to the composite risk score (ICC =.97) and the risk rating (ICC =.88).
•Lodewijks et al. (2008a) the SAVRY demonstrated similar ICC in relation to the summary risk scale score (.82).
•Inter rater reliability was measured by Shepherd et al. (2014) using twenty eight cases from a sample of 213 adolescents in Australia. The ICC level was almost perfect at .97, identifying the level of agreement between the two raters.
•Shepherd et al. (2014) found that the SAVRY had the ability to identify specific treatment targets for youth, suggesting that the tool is able to link dynamic social and environmental factors with reoffending outcomes.
•Using a sample of 145 Spanish juveniles, Hilterman et al. (2014) found that the ICCs for general and violent recidivism were good (.66) and excellent (.76) respectively. The ICCs for subscales and total scores also ranged from 0.60 to 0.89, falling within good and excellent levels.
•Penney et al. (2010) the SAVRY demonstrated an ICC of .91 for the composite score.
Inter Rater Reliability
b)International Research
•Selby (2018) looked at the inter rater reliability of the SAVRY amongst mental health professionals, looking at professional characteristics like perception of their confidence and objectivity in ratings. Self reported confidence was not associated with increased reliability in scoring, suggesting a need for training.
Structure of the SAVRY is modelled on other existing guided assessment protocols such as the HCR 20.The item content is focused specifically on the risk in adolescents. The 24 risk items have been drawn from literature and research on adolescent development and violence in youth (Borum et al., 2006).
•Spice et al. (2009) the SAVRY composite score significantly predicted adult sentencing and/or transfer to courts in a sample of 74 adolescents (AUC = .71)
•Lodewijks et al. (2008) found moderate to large AUC values found for various types of disruptive behaviours including physical violence (.86), violence against objects (.74) and verbal abuse (.74). The composite score and summary risk rating were significantly above chance prediction of future violence.
•Viljoen et al. (2008) found the SAVRY composite score was able to predict non sexual aggression during treatment (AUC = .69) and post discharge (AUC = .77). It could not, however, significantly predict sexual aggression during treatment or sexual offences post discharge.
•Penney et al. (2010) found that few youth within a high risk sample demonstrated any protective factors as per the SAVRY. This led the researchers to suggest that the protective factor items on the SAVRY are perhaps not fully measuring strengths in high risk adolescents.
General Predictive Accuracy
•To measure risk over time, Vilijoen et al. (2017) carried out 508 risk assessments on 146 adolescents every three months for a year. This created partial support for the ‘internal sensitivity’ of the SAVRY in measuring changes in risk over time, with a modest proportion of youth displaying reliable changes. The link between change scores and reoffending (external sensitivity) was moderately supported by the results.
a)UK Research
•McGowan et al. (2011) found good predictive accuracy (AUC = .72) in correctly identifying violent youths upon carrying out a retrospective file review on 87 adolescents (aged 12 18) in educational settings
None available at present.
•Welsh et al. (2008) significant ROC values in relation to the prediction of general (.77) and violent recidivism (.81).
b)International Research
•Singh, Grann and Fazel (2011) in a meta analysis, the SAVRY achieved a median AUC value of .71 in predicting violent recidivism.
RATED page updated: July 2019 © Risk Management Authority 2019
Validation History
•Perrault et al. (2017) found that the SAVRY completed by juvenile probation officers in a sample of 383 adolescents significantly predicted violent reoffending with an AUC of 0.69.
•Childs and Frick (2016) found that the SAVRY yielded similar measures of risk across age groups of 13 to 15 and 16 to 18.
•A study by Childs et al. (2013) provided moderate support for the predictive validity of the SAVRY in a sample of 158 adjudicated youth.
•For longer term follow up periods of four to seven years, the SAVRY was shown to predict violence in adolescents (Sijitsema et al., 2015).
RATED page updated: July 2019 © Risk Management Authority 2019
•Lawing et al. (2017) found that SAVRY was able to both distinct violent from non violent offending in a sample of 505 adolescents and predict violent and non violent recidivism over a year follow up period. The ‘anger control’ item was found to be an important indicative factor for risk.
•Chu et al. (2016) applied the SAVRY to 165 adolescents and discovered that the total scores were moderately predictive of violent and general recidivism with AUCs of .65 and .72 respectively. The Protective score of the SAVRY also generated moderate and large predictive accuracy for violent and general recidivism with AUCs of .69 and .72.
•The predictive validity of the SAVRY was .75 for reoffending; although there was not any predictive validity on the protective factors (Hilterman et al., 2014).
•Ortega Campus et al. (2017) found the SAVRY differentiated between adolescents at low and high risk of reoffending and showed good predictive capacity with an AUC of 0.737 for risk total score and an AUC of 0.748 for the summary risk rating.
•The summary risk rating of the SAVRY was a significant predictor of serious violence in a sample of 56 adolescents in Sweden with an AUC of .80 (Åström et al., 2015).
•A study of 213 adolescents found the SAVRY was able to predict general recidivism (AUC=.71) (Shepherd et al., 2014).
•Soderstrom, Childs and Frick (2019) utilised the SAVRY to analyse the impact of protective factors on reoffending using a sample (n=460) of post adjudication juveniles in a Southern state. Findings indicated that protective factors did not predict reoffending when controlling for risk domains. It was found, however, that certain protective factors buffer the effect of some of the risk domains.
•Vilijoen et al. (2018) examined the predictive validity of the SAVRY for 216 adolescents on probation. AUCs generated for violent charges were .66 and .60 for total score and summary risk ratings respectively. For all charges, the validity was slightly lower, with AUCs of .63 and .59 for total score and summary risk ratings respectively.
•Schmidt et al. (2011) low to moderate predictive accuracy observed between non violent (AUC =.68) and violent (.57) recidivism in relation to the SAVRY composite score.
None available at present.
•Testing the SAVRY on 100 male juvenile who had committed sexual offences found that the total score and overall risk rating significantly predicted general and non sexual recidivism (AUC=>66 and .64 respectively) (Owens, 2011).
a)UK Research
b) International Research
RATED page updated: July 2019 © Risk Management Authority 2019
Applicability: Females
Validation History
•Hilterman et al. (2018) conducted a longitudinal study of 5205 male juveniles through the Catalan justice system from 2006 2014 to test the ability of the SAVRY to measure distinct change over time. Results showed that the tool might not be sufficiently sensitive to measure changes in juveniles who offend over time.
•A study in China found that the AUC for the total risk score was predictive at 0.68. The protective factors, however, yielded an AUC at 0.60, which is lower than it tends to be for those who offend in Western countries. This led the authors to suggest that the cultural factors relative to China may not be measured with the SAVRY protective items (Zhou et al., 2017).
b)International Research
•Vincent et al. (2012) found that ethnicity moderated the association between summary risk ratings on the SAVRY and re arrests within a 1 5 year follow up. For instance, White individuals with moderate to high summary risk ratings were almost 4.5 times more likely to be re arrested for a non violent offence than those of other ethnic origins.
Validation History
•Penney et al. (2010) found moderate to high predictive accuracy between the composite score and violent (AUC = .72) and non violent (AUC =.65) recidivism.
Applicability: Mental Disorders
Validation History
•A UK study of 76 male youth with conduct disorder (CD) and 33 with conduct disorder and attention deficit hyperactivity disorder (CD/ADHD) found that the CD/ADHD group had higher scores on the SAVRY on the social and individual domains. The SAVRY showed more
•Lodewijks et al. (2008a) the SAVRY demonstrated predictive accuracy in a sample of females (AUC =.85). In spite of this, there was a higher rate of false positives in females who offended than males.
Applicability: Ethnic Minorities
a)UK Research
None available at present.
•Childs and colleagues (2013) used administrative data from 292 adjudicated juveniles placed in state custody to test the SAVRY across genders. Results support the use of the SAVRY for both boys and girls.
a)UK Research
•Dolan and Rennie (2008) found moderate AUC values for the composite SAVRY score and violent (.64) and general recidivism (.69) in a sample of males diagnosed with conduct disorder.
RATED page updated: July 2019 © Risk Management Authority 2019
•Meyers and Schmidt (2008) found moderate to high accuracy in predicting violent recidivism in Native Canadian youth at 1 (AUC =.64) and 3 year (AUC =.84) follow up periods.
•The SAVRY can aid assessors in identifying risk and responsivity factors specific to the individual (e.g. negative attitudes, low empathy).
Contribution to Risk Practice
•The dynamic factors included in the SAVRY can act as targets for change.
•McLachlan et al. (2018) carried out research into the predictive validity of the SAVRY in youth with ‘foetal alcohol spectrum disorder,’ using a sample of 50 youth with this condition and 50 without FASD or prenatal alcohol exposure. The SAVRY was shown to predict recidivism in this offending population.
b)International Research
•Parmar (2016) found that those who scored at the moderate and high levels in the SAVRY had significant mood disturbances and feelings of loneliness and hopelessness. It was thus suggested that the SAVRY should be used in routine psychiatric assessments to identify youths at risk of violence and allow for treatment strategies to be devised.
•The implementation of the SAVRY in a probation office led to a reduction in both secure and non secure placement rates and the use of maximum and intensive supervision (Vincent et al., 2012).
•The tool identifies risk, responsivity and protective factors that could contribute to risk management strategies such as victim safety planning and risk scenario planning.
•Childs et al. (2013) suggest that a focus on non violent delinquency risk coupled with risk of violence could increase the usefulness of the SAVRY in devising management and intervention strategies for non violent or low risk individuals on probation.
•The SAVRY can be time consuming to administer.
predictive accuracy of violent reoffending for the CD group (Khanna et al., 2014).
•A study of adolescents with mental disorders carried out in a psychiatry setting in Finland found that the summary risk rating of the SAVRY was the most accurate predictor of violent offending as well as non violent criminal conduct (Gammelgard et al., 2015).
RATED page updated: July 2019 © Risk Management Authority 2019
•It was found that the inclusion of the SAVRY and structured case plans led to significantly better case plans for 216 adolescents on probation (Vilijoen et al., 2018)
Other Considerations
•A doctoral dissertation examined whether file only raters can reliably and accurately code the SAVRY in cases where standard administration is not possible. Findings indicated that to reliably score the SAVRY solely with file information, the evaluator must have access to an adequate source of information on the defendant. It is suggested that there may be a threshold level of data to allow for the SAVRY to be accurately coded (Burl, 2012).
•The instrument is able to be used by adolescents in hospital, mental health and justice settings (Sher et al., 2017; Viljoen et al., 2012a).
Strengths
Year 2010 Description
Assessor Qualifications
•The START:AV was found to have strong current validity with the SAVRY and identify a greater number of strengths (Viljoen et al., 2012a).
•This is an SPJ instrument focusing on assessing short term risk (up to three months) and strength factors in adolescents. All items are potentially dynamic in nature (Singh et al., 2014).
•This is an adolescent version of the START risk assessment tool. Developers of START worked alongside individuals with clinical and research expertise in managing adolescents to develop the START:AV. It was developed out of a need to address factors like self harm, suicide, victimisation and substance abuse in risk assessment (Vilijoen et al., 2012b).
The recommendations provided in the START:AV User Guide are to obtain formal training via a workshop if possible, study the User Guide and companion Knowledge Guide, establish competency through a minimum of three practice cases and regularly refresh knowledge about the tool.
•The START:AV is said to complement other risk measures in a number of ways: examination of the broader adverse outcomes that adolescents are vulnerable to; offers a balanced overview of strengths and vulnerabilities; focuses on dynamic factors that are relevant to short term risk (Viljoen et al., 2012a).
•It has been suggested that the START:AV may be used to classify dynamic factors as acute or stable, which could be useful in identifying treatment options and interventions (Sellers et al., 2017).
Age Appropriateness
Category Youth Assessment: Violence Risk (Validated)
RATED page updated: July 2019 © Risk Management Authority 2019
•It consists of dynamic and protective factors that are rated from 0 to 2 for their presence within the review period. Risk estimates of low, moderate or high are given on eight outcome domains: violence, self harm, suicide, unauthorised absence, substance abuse, self neglect, victimisation and general offending (Sellers et al., 2017; Sher et al., 2017).
Author / Publisher Viljoen, Nicholls, Cruise, Desmarais and Webster
Name of Tool Short Term Assessment of Risk & Treatability: Adolescent Version (START:AV)
12 18
•Viljoen et al. (2012a) found that ICCs were in the good to excellent range, with any disagreements relating to low/moderate and moderate/high risk.
Applicability: Females
Validation History
Validation History
b)International Research
a)UK Research
Empirical Grounding
General Predictive Accuracy
An extensive literature review was undertaken by the authors to formulate risk and protective factors for adolescents. All of the items in the adult version of START were found to be relevant to young people, so these were retained in the START:AV. General Offending was added as an outcome and detailed coding instructions were provided to explain how risk and protective factors could manifest in adolescents: parenting and home environment, as well as relationships with caretakers and peers (Viljoen et al., 2012b).
Sher et al. (2017) found there were gender differences in predictive validity, with no significant relationships being found when it was applied to a female sample. It is, therefore, suggested that the START:AV items do not accurately reflect the strengths and vulnerabilities specific to female self harm and aggression.
RATED page updated: July 2019 © Risk Management Authority 2019
Inter-Rater Reliability
None available at present.
b)International Research
•Inter rater agreement was evident (k>.67) in 10% of randomly selected cases in a study by Singh et al. (2014)
Validation History
a)UK Research
•In a sample of 90 adolescents, Viljoen et al. (2012a) found that START:AV risk estimates and vulnerability total scores predicted a number of adverse outcomes: violence, offending, victimisation, suicidal ideation and substance abuse.
•In a study by Sher et al. (2017), the START:AV total vulnerabilities and verbal aggression and the total vulnerabilities and physical aggression scores yielded a moderate to large effect size.
Applicability: Ethnic Minorities
•Sher and Gralton (2014) surveyed staff members in a UK based medium secure service for adolescents to determine their views about the START:AV. Findings showed that staff members felt the instrument was straightforward to use, although there were difficulties in completing risk formulation and making distinctions in ratings.
•Sher et al. (2017) carried out a study within a medium secure adolescent service with a sample divided between those on pathways for mental disorder and developmental disabilities (individuals with a diagnosis of a learning disability or autism spectrum disorder). The study found there was evidence for the predictive validity of START:AV in male adolescents with and without developmental disabilities. Predictions for property damage, physical and verbal aggression were significant for the non developmental disabilities group
None available at present.
•De Beauf, de Vogel and de Ruiter (2019) assessed the implementation of the START:AV in a residential youth care facility in the Netherlands. The majority of staff members perceived the START:AV core constructs as useful for treatment and the completion rate for assessments was acceptable. A lack of integration into clinical case conferences and increased workload, however, meant that satisfaction with the tool decreased for staff members over time.
•Viljoen (2014) applied the START:AV retrospectively to a group of 30 American Indian and Alaska Native youth in a residential centre in the United States. Vulnerability and strength scores were predictive of violence with AUCs of .78 and .67 respectively.
Contribution to Risk Practice
Validation History
Applicability: Mental Disorders
b) International Research
a)UK Research
Other Considerations
•Rather than relying solely on the START:AV, Viljoen et al. (2012b) recommended that it should be supplemented with additional evidence based approaches.
RATED page updated: July 2019 © Risk Management Authority 2019
b)International Research
•Singh et al. (2014) found there were discrepancies between START:AV assessments and treatment plans, for adolescents with higher vulnerabilities ratings (particularly females) had fewer interventions targeting their specific needs. This elucidates the need for interventions to be tailored to risk assessment scoring.
None available at present.
a)UK Research
RATED page updated: July 2019 © Risk Management Authority 2019
•Training for this tool is available online or the possibility of a START:AV author travelling to venues to provide in person training may also be considered.
•The SAPROF only measures protective factors grouped into four domains: resilience, consisting of social competence, coping, self control and perseverance; motivational items of future orientating attitude towards agreements and conditions, medication, school work and leisure activities; relational domain of parents/guardians, peers and other supportive relationships; external items, containing pedagogical climate, professional care and court order.
Assessor Qualifications
•Experience and training in conducting individual assessments with adolescents.
The18authors
•Every item is rated on a seven point scale and the strength of this over the course of six months. The rater also highlights the critical factors to preventing violent offending for each individual and these are then incorporated into treatment targets.
•Experience and training in the administration and interpretation of tests and semi structured interviews.
© Risk Management Authority 2019
Name of Tool Structured Assessment of Positive Factors for Violence Risk: Youth Version (SAPROF: YV)
•Zeng et al. (2015) applied the SAPROF to 97 Singaporean youth and found that the total domain score did not result in acceptable predictive validity. It was, thus, advocated that a youth version of the tool should be developed and used for testing adolescents.
Age Appropriateness
Author / Publisher de Vries Robbé, Geers, Stapel, Hilterman and de Vogel
12
maintain that it could possibly be used up to age 23.
•It is to be used in addition to other tools like the SAVRY and YLS/CMI to formulate risk scenarios and calculate an overall risk score consisting of both risk and protective factors.
•The SAPROF: YV is an adolescent version of the SAPROF.
•Assessors should also be familiar with the most recent professional and research literature on the causes and prediction of violence in youth.
Year 2015 Description
•Two pilot studies were carried out on the SAPROF:YV. Validation studies are said to be on going in European, Asian and North American countries.
•An unpublished Masters dissertation examined aggression in 69 adolescents using the SAPROF:YV and SAVRY. The SAPROF:YV displayed good convergent and discriminant validity with
RATED page updated: July 2019
Category Youth Assessment: Violence Risk (Awaiting Validation)
Tool Development
the SAVRY. The SAPROF: YV predicted the absence of verbal and physical aggression; it was also found to be better at predicting higher risk adolescents than lower risk ones. The SAPROF:YV did not, however, add incrementally to the SAVRY risk scores (Bhanwer, 2016).
•The SAPROF:YV was implemented nation wide across the Netherlands in juvenile justice institutions, with the requirement that this is coupled with the SAVRY in assessments (de Vries Robbe and Willis, 2017).
•It is available in English, Dutch and Spanish.
RATED page updated: July 2019 © Risk Management Authority 2019
General Notes
•Further information may be found at http://www.saprof.com/saprof youth version/ or by emailing saprof yv@saprof.com.
Description
Tool Development
•It is a 23 item clinician rated risk assessment and treatment planning tool designed for the assessment of risk, need, responsivity and treatment change for youth at risk of committing violent offences. It consists of static and dynamic items.
Assessor Qualifications
•Stockdale (2008) found the tool demonstrated reasonable inter rater reliability for the composite score (ICC = .90). The VRS:YV also displayed large AUC scores for previous offending behaviours; violent offences and any prior offences (AUCs = .70 and .74 respectively).
•Change is assessed using an adapted version of the process of change model (Prochaska et al., 1992).
Category Youth Assessment: Violence Risk (Awaiting Validation)
•The VRS:YV is scored on a 4 point ordinal scale from 0 to 3. Those that are rated 2 or 3 are considered criminogenic, thus are given priority for services. The tool is unique for including a scheme to assess the readiness for treatment.
Adolescents within the youth justice system.
Author / Publisher Wong, Lewis, Stockdale and Gordon Year 2003 2010
Name of Tool Violence Risk Scale: Youth Version (VRS:YV)
Assessors should undertake the relevant training and present with understanding adolescent development (e.g., from educational and/or employment experiences).
Age Appropriateness
•The VRS:YV is the adolescent version of the VRS and is closely modelled on it.
•A review of the relevant files and a semi structured interview with the individual are required to rate items on the VRS:YV.
•High ratings on this measure indicate increased risk for violence.
•The tool is a youth adaptation of the adult VRS. It was developed by reviewing the relevant literature on violence risk assessment and treatment in youth. Particular attention was heeded to factors specific to young people: the importance of family, parents, peers and school overall in their lives (Stockdale et al., 2014).
•VRS:YV demonstrated high inter rater reliability (ICC = .90) for the composite pre treatment total (i.e. static and dynamic scores). Its total scores significantly predicted violent (AUC = .77), nonviolent (AUC = .72), and general (AUC = .73) recidivism over an average 7 year follow up. VRS:YV also predicted youth and adult violence (AUC = .75 and .73, respectively). VRS:YV predicted future violence among female (AUC = .66) and Indigenous (AUC = .72) youth. This preliminary research is
RATED page updated: July 2019 © Risk Management Authority 2019
•The 2014 publication by Stockdale et al. extends the robust findings from the dissertation with an extended follow up The inter rater reliability was found to be excellent with an ICC of .90. The static, dynamic and total scores were all significant predictors of recidivism with an AUC range of .65 to .78.
RATED page updated: July 2019 © Risk Management Authority 2019
•Validation for this tool is still in its preliminary stages.
based on the dissertation of Stockdale (2008) and incorporates a longer term follow up and more comprehensive outcome data.
•Stockdale et al. (2014) found that the tool demonstrated moderate to high predictive accuracy for violent and general recidivism in males, females, and Indigenous youth across developmental subgroups.
•For further information about the VRS YV, please email Keira Stockdale: keira.stockdale@police.saskatoon.sk.ca
General Notes
RATED page updated: July 2019 © Risk Management Authority 2019
Borum, R., Bartel, P. and Forth, A. (2006) Manual for the Structured Assessment for Violence Risk in Youth (SAVRY). Odessa, FL: Psychological Assessment Resources. [Not accessible]
Borum, R., Lodewijk, H., Bortel, P. A. and Forth, A. E. (2010) ‘Structured Assessment of Violence Risk in Youth (SAVRY).’ In R. K. Otto and K. S. Douglas. Handbook of Violence Risk Assessment. New York, London: Routledge, 63 80. Access here
Childs, K. K. and Frick, Paul J. (2016) ‘Age differences in the Structured Assessment of Violence Risk in Youth (SAVRY).’ International Journal of Forensic Mental Health 15(3), 211 221. Access Here.
Burl, J. D. (2012) ‘Are file review based SAVRY ratings of violence risk reliable?’ Unpublished doctoral thesis. Philadelphia, PA: Drexel University. Access Here.
Chu, C. M., Goh, M. L. and Chong, D. (2016) ‘The Predictive Validity of SAVRY ratings for assessing youth offenders in Singapore.’ Criminal Justice & Behavior 43(6), 793 810. Access Here
Dolan, M. C. and Rennie, C. E. (2008). ‘The Structured Assessment of Violence Risk in Youth as a predictor of recidivism in a United Kingdom cohort of adolescent offenders with conduct disorder.’ Psychological Assessment 20(1), 35 46. Access here.
Gammelgard, M., Kovisto, A. M., Eroken, M. and Kaltiala Heino, R. (2015) ‘Predictive validity of the structured assessment of violence risk in youth: a 4 year follow up.’ Criminal Behavior and Mental Health 25(3), 192 206. Access Here
Khanna, D., Shaw, J., Dolan, M. and Lenox, C. (2014) ‘Does diagnosis affect the predictive accuracy of risk assessment tools for juvenile offenders: conduct disorder and attention deficit hyperactivity disorder.’ Journal of Adolescence 37(7), 1171 1179. Access Here.
Lawing, K., Childs, K. K., Frick, P. J. and Vincent, G. (2017) ‘Use of Structured Professional Judgement by probation officers to assess risk for recidivism in adolescent offenders.’ Psychological Assessment 29 (6), 652 663. Access Here
Åström,SAVRY
Hilterman, Ed L. B., Bongers, Ilja L., Nicholls, Tonia L. and van Nieuwenhuizen, Chris. (2018) ‘Supervision trajectories of male juvenile offenders: growth mixture modelling on SAVRY risk assessments.’ Child and Adolescent Psychiatry and Mental Health 12(5). Access Here
Hilterman, E. L. B., Nicholls, T. L. and van Nieuwenhuizen, C. (2014). ‘Predictive validity of risk assessments in juvenile offenders: Comparing the SAVRY, PCL:YV, and YLS/CMI with unstructured clinical assessments.’ Assessment 21(3), 324 339. Access here.
Childs, Kristina K. Ryals, Jr., J., Frick, Paul J., Lawing, P. Phillippi, S. W. and. Deprato, D. K. (2013) ‘Examining the validity of the structured assessment of violence risk in youth (SAVRY) for predicting probation outcomes among adjudicated juvenile offenders.’ Behavioural Science and the Law 31, 256 270. Access Here
T., Gumpert, C. H., Andershed, A. K. and Forster, M. (2017) ‘The SAVRY Improves Prediction of Reoffending: A Naturalistic Longitudinal Comparative Study.’ Research on Social Work Practice 27(6), 683 694. Access Here.
VALIDATED TOOLS
McGowan, M. R., Horn, R. A. and Mellott, R. N. (2011). ‘The predictive validity of the Structured Assessment of Violence Risk in Youth in secondary educational settings.’ Psychological Assessment 23(2), 478 486. Access here.
Sijitsema, J. J., Kretschmer, T. and van Os, T. (2015) ‘The structured assessment of violence risk in youth in a large community sample of young adult males and females: the TRAILS study.’ Psychological Assessment 27 (2), 669 677. Access Here
RATED page updated:
Shepherd, S. M., Luebbers, S., Ogloff, J. R. P. Fullam, R. and Dolan, M. (2014) ‘The Predictive Validity of Risk Assessment Approaches for Young Australian Offenders.’ Psychiatry, Psychology and Law 21(5), 801 817. Access here.
McLachlan, Kaitlyn, Gray, Andrew L., Roesch, Ronald, Douglas, Kevin S. and Viljoen, Jodi L. (2018) ‘An evaluation of the predictive validity of the SAVRY and the YLS/CMI in justice involved youth with fetal alcohol spectrum disorder.’ Psychological Assessment 30(12), 1640 1651. Access Here.
Lodewijks, H. P. B. de Ruiter, C. and Doreleijers, T. H. A. (2008) ‘Gender Differences in Violent Outcome and Risk Assessment in Adolescent Offenders After Residential Treatment.’ International Journal of Forensic Mental Health 7(2), 133 146. Access here
Parmar, A. (2016) ‘Structured Assessment for Violence Risk in Youth (SAVRY) Findings in High Risk School Children and Adolescents.’ Journal of the American Academy of Child and Adolescent Psychiatry 55 (10), S13. Access Here
Penney S. R., Lee Z., Moretti M. M. (2010) ‘Gender differences in risk factors for violence: An examination of the predictive validity of the Structured Assessment of Violence Risk in Youth.’ Aggressive Behavior 36(6), 390 404. Access here.
Perrault, R. T., Vincent G. M. and L. S. Guy. (2017) ‘Are risk assessments racially biased? Field study of the SAVRY and YLS/CMI in probation.’ Psychological Assessment 29 (6), 664 678. Access Here
Schmidt, F., Campbell, M. A., and Houlding, C. (2011) ‘Comparative Analyses of the YLS/CMI, SAVRY, and PCL:YV in Adolescent Offenders: A 10 year Follow Up Into Adulthood.’ Youth Violence and Juvenile Justice 9(1), 23 42. Access Here.
July 2019 © Risk Management Authority 2019
Selby, Sarah Elizabeth. (2018) Inter rater reliability of the Structured Assessment of Violence Risk in Youth (SAVRY) amongst mental health professionals. Doctoral thesis in Clinical Psychology. Glasgow: Institute of Health and Wellbeing, University of Glasgow. Access Here
Meyers, J. R. and Schmidt, F. (2008) ‘Predictive Validity of the Structured Assessment for Violence Risk in Youth (SAVRY) With Juvenile Offenders.’ Criminal Justice and Behavior, 35(3), 344 355. Access Here
Ortega Campus, E., Garcia Garcia, J. and Zaldicor Basorto, F. (2017) ‘The Predictive Validity of the Structured Assessment of Violence Risk in Youth for Young Spanish Offenders.’ Frontiers in Psychology 8 (577), 1 9. Access Here.
Owens, T. (2011) The Utility of the SAVRY in Predicting Recidivism Among Juvenile Sex Offenders. Unpublished Psychology Thesis. Bristol, RI: Roger Williams University. Access Here.
Zhou, J., Witt, K., Cao, X. Chen, C. and Wong, X. (2017) ‘Predicting Reoffending using the Structured Assessment of Violence Risk in Youth (SAVRY): A 5 year follow up study of male juvenile offenders in Hunain Province, China.’ PLOS One 12 (1), 1 11. Access Here
Sellers, B. G., Desmarais, S. L. and Hanger, M. W. (2017) ‘Measurement of Change in Dynamic Factors using the START: AV.’ Journal of Forensic Psychology Research and Practice 17(3), 198 215. Access Here
Vilijoen, J. L., Muir, C. S., Muir, N. M., Cochrane, D. M. and Brodersen, E. M. (2018) ‘Improving Case Plans and Interventions for Adolescents on Probation: The Implementation of the SAVRY and a Structured Case Planning Form.’ Criminal Justice and Behavior 46(1), 42 62. Access Here
Yates, H. M. (2005) A Review of Evidence Based Practice in the Assessment and Treatment of Sex Offenders. U.S.: Office of Planning, Research, Statistics nad Grants, Pennsylvania Departments of Corrections. Access here.
Scalora, M., Cuadra, L., Bader, S., Chávez, V., Ullman, D., and Lawrence, L. (2008) ‘Assessing Risk for Violence in Adolescents Who Have Sexually Offended: A Comparison of the J SOAP II, J SORRAT II, and SAVRY.’ Criminal Justice and Behavior 35(1), 5 23. Access Here.
Soderstrom, M. F. P., Childs, K. K. and Frick, P. J. (2019) ‘The Role of Protective Factors in the Predictive Accuracy of the Structured Assessment Violence Risk in Youth (SAVRY).’ Youth Violence and Juvenile Justice. Access Here.
Vincent, Gina M., Guy, L. S., Geshenson, B. G. and McCabe, P. (2012) ‘Does risk assessment make a difference? Results of implementing the SAVRY in juvenile probation.’ Behavioural Sciences and the Law 30(4), 384 405. Access Here.
Vilijoen, J. L., Shaffer, C. S., Gray, A. L. and Douglas, K. S. (2017) ‘Are adolescent risk assessment tools sensitive to change? A framework and examination of the SAVRY and the YLS/CMI.’ Law and Human Behavior 41 (3), 244 257. Access Here
RATED page updated: July 2019
© Risk Management Authority 2019
Spice, A. (2009) Psychological assessment for adult sentencing of juvenile offenders: an evaluation of the RSTI and the SAVRY. Unpublished Masters thesis. Smino Fraser University. Access Viljoen,HereJ.L.,
Singh, J.P., Grann, M. and Fazel, S. (2011) ‘A comparative study of violence risk assessment tools: a systematic review and meta regression analysis of 68 studies involving 25,980 participants.’ Clinical Psychology Review 31(3), 499 513. Access here.
Welsh, J. L., Schmidt, F., McKinnon, L., K., H. and Meyers, J. R. (2008) ‘A Comparative Study of Adolescent Risk Assessment Instruments: Predictive and Incremental Validity.’ Assessment 15(1), 104 115. Access Here
DeSTART:AVBeuf,T. L. F., de Vogel, V. and de Ruiter, C. (2019) ‘Implementing the START:AV in a Dutch residential youth facility: Outcomes of success.’ Translational Issues in Psychological Science 5(2), 193 205. Access Here
TOOLS AWAITING VALIDATION
Viljoen, J. L., Cruise, K. R., Nicholls, T. L., Desmarais, S. L. and Webster, C. D. (2012a) ‘Taking stock and taking steps: the case for an adolescent version of the short term assessment of risk and treatability.’ International Journal of Forensic Mental Health 11, 135 149. Access Here.
Prochaska, J. O., DiClemente, C. C. and Norcross, J. C. (1992) ‘In search of how people change: Applications to addictive behaviors.’ American Psychologist 47(9), 1102 1114. Access here.
VRS:YV
RATED page updated: July 2019 Management Authority 2019
Webster, C. D. and Belisle, E. (2014) ‘How literature can add value to structured professional judgments of violence risks: an illustrative rare risk example inspired by Alice Munro’s Child Pay.’ Archives of Forensic Psychology 1(1), 14 26. [Not accessible]
Stockdale, K. C., Olver, M. E. and Wong, S. P. C. (2014) ‘The validity and reliability of the Violence Risk Scale Youth Version in a Diverse Sample of Violent Young Offenders.’ Criminal Justice and Behaviour 41 (1), 114 138. Access here
Viljoen, S. (2014) ‘Using strengths based measures to assess and manage risk of future negative outcomes.’ New Mexico: Indian Health Service Clinical Rounds. Access Here
Zeng, G., Chu, C. M. and Lee, Y. (2015) ‘Assessing Protective Factors of Youth with Sexually Offenders in Singapore: Preliminary Evidence on the utility of the DASH 13 and the SAPROF.’ Sexual Abuse: A Journal for Research and Treatment 27 (1), 91 108. Access here
Sher, M. A. and Gralton, E. (2014) ‘Implementation of the START:AV in a secure adolescent service.’ Journal of Forensic Practice 16 (3), 184 193. Access Here.
© Risk
Viljoen, J. L., Benefteau, J. L.. Gulbransen, E. Brodersen, E., Desmarais, S. L., Nicholls, T. L. and Cruise, K. R. (2012b) ‘Assessment of multiple risk outcomes, strengths and change in the START: AV: a short term protective study with adolescent offenders.’ International Journal of Forensic Mental Health 11 (3), 165 180. Access Here.
Stockdale, K. C. (2008). The validity and reliability of the Violence Risk Scale Youth Version (VRS YV). (Unpublished doctoral dissertation). Saskatoon, Canada: University of Saskatchewan. Access here
de Vries Robbe, M. and Willis, G. M. (2017) ‘Assessment of Protective Factors in clinical practice.’ Aggression and Violent Behaviour 32, 55 63. Access here.
SAPROF: Youth Version Bhanwer, A. K. (2016) ‘The Structured Assessment of Protective Factors for Violence Risk Youth Version (SAPROF YV): The Association Between Protective Factors and Aggression in Adolescents.’ Masters of Arts Thesis. Burnaby, Canada: Simon Fraser University. Access here
Sher, M. A., Warner, L., McLean, A., Rowe, K. and Gralton, E. (2017) ‘A prospective validation study of the START:AV.’ Journal of Forensic Practice 19(2), 115 129. Access Here.
Singh, J. P., Desmarais, S. L., Sellers, B. G., Hylton, T., Torotti, M. and van Dorn, R. A. (2014) ‘From risk assessments to risk management: matching interventions to adolescent offenders strengths and vulnerabilities.’ Child Youth Service Review 47(1), 1 9. Access Here
Age Appropriateness
Description
•The ERASOR is based on the structured professional approach and, as such, does not apply cut off scores or formulas in determining the individual’s level of risk.
Assessor Qualifications
•The ERASOR is a 25 item structured assessment tool that is designed to assess the risk of sexual recidivism in adolescents who have committed prior sexual offences.
Strengths
RATED page updated: August 2019 © Risk Management Authority 2019
a)UK Research
Category Youth Assessment: Sexual Violence Risk (Validated)
Inter Rater Reliability
•The items are clustered under five subscales; (1) sexual interests, attitudes and behaviours, (2) historical sexual assaults, (3) psychosocial functioning, (4) family/environmental functioning and (5)treatment. All risk factors are coded as either Present, Possibly Present, Not Present or Unknown.
Year 2001
•Considers factors relevant to treatment interventions.
Empirical Grounding
Name of Tool Estimate of Risk of Adolescent Sexual Offence Recidivism (ERASOR)
Assessors must possess the relevant training/experience in youth assessment
The authors used three sources of information when establishing the items found on the ERASOR published studies of adolescent sexual offence recidivism (10 studies), published guidelines of clinical judgement of risk and protective factors, and literature on adult sexual offending behaviour (Worling, 2004).
Author / Publisher Worling and Curwen
12 18
No empirical evidence at present.
RATED page updated: August 2019 © Risk Management Authority 2019
•Nelson (2011) reported an ICC value of .64 for the total score.
•Viljoen et al. (2009) the ERASOR demonstrated an ICC of .90 for the total score and .75 for the clinical risk rating.
•Edwards and colleagues (2005) found that kappa levels ranged from fair to excellent for the different ERASOR domains: attitudes supportive of sexual offending .44; interpersonal aggression .79; unwilling to alter sexual interests/attitudes .82; impulsivity, .88 and ever a male victim 1.0).
•In her doctoral dissertation, Skowron (2004) calculated inter rater reliability for 16% of the sample. The total score for the ERASOR was .87. All the scales on the ERASOR had significant ICC: psychosocial functioning (ICC=.87); historical sexual assaults (ICC=.78); sexual interests, attitudes and behaviours (ICC=.74); family/environmental functioning (ICC=.73); treatment (ICC=.55).
•A doctoral dissertation found that IRR for clinical judgment was significant at .86 (Chávez, 2010).
•Chu et al. (2011) found fair inter rater reliability for the ERASOR total score (ICC = .49) and clinical risk rating (ICC=.43).
•In an unpublished Master’s thesis, Morton (2003) examined the ICC of the clinical judgment risk rating (.68) and total score (.94) on the ERASOR.
b)International Research
•Worling, Bookalam, and Littlejohn (2012) found excellent ICC value of .88 for the ERASOR composite score.
•Hersant’s (2006) doctoral dissertation found that the ERASOR total score was .87.
•Rajlic and Gretton (2010) reported an ICC value of .89 for the composite score and .78 for the clinical risk rating. ERASOR risk categories were also examined: sexual interests, attitudes and behaviours (ICC=.74), historical sexual assaults (ICC=.78), psychosocial functioning (ICC=..87), family environmental functioning (ICC=..73) and treatment (ICC=..55).
•In an unpublished doctoral dissertation, McCoy (2007) found that the IRR for the ERASOR total score was .87.
a)UK Research
•Worling, Bookalam and Littlejohn (2012) moderate to high AUC values were observed for the composite ERASOR score in the prediction of sexual (.72), and non sexual violent recidivism (.65). Although the measure was unable to predict non violent recidivism. In shorter follow up period (2.5 years), the composite score achieved an AUC value of .93 in a sub group of 70 individuals who had offended.
General Predictive Accuracy
b)International Research
•Viljoen et al. (2009) the ERASOR composite score did not significantly predict sexual, non sexual and violent recidivism when applied to 193 adolescent males. The clinical risk rating was moderately predictive of sexual recidivism (AUC=.64).
•Skowron (2004) the tool demonstrated predictive accuracy in predicting sexual recidivism (AUC = .71).
•Rojas Mejia (2013) found the IRR was fair for clinical risk rating (.42) and good for the total score (.71).
•Nelson (2011) found that the clusters of items ranged from very poor (.03) to excellent (.93). Inter rater reliability was .76 for the total score and .64 for the clinical judgment rating.
RATED page updated: August 2019 © Risk Management Authority 2019
•When applied to a sample of 597 male juveniles with sexual offences, the ERASOR was best suited to predict sexual recidivism with 0.5 to 3 years (Barra et al., 2018).
Validation History
No empirical evidence at present.
•Rajlic and Gretton (2010) the ERASOR demonstrated moderate predictive accuracy in relation to sexual (AUC =.71) and non sexual (AUC =.71) recidivism. Clinical judgment ratings were significantly predictive of sexual reoffending (AUC=.67).
•In a systematic review of studies, Campbell and colleagues (2016) found evidence that the ERASOR could assist in the predict of risk: three studies recording AUCs of .71, .72 and .77; although one found it did not significantly predict sexual recidivism with an AUC of .54. The ERASOR may also be able to predict future non sexual
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
recidivism but the effect is not consistent across all studies.
No empirical evidence at present.
•An unpublished doctoral dissertation by Hersant (2006) applied the ERASOR to 91 adolescent males. Findings showed that the total score (AUC=.66) and clinical judgment risk ratings (AUC=.66) were able to significantly differentiate those adolescents who reoffended from those who offended for the first time.
•A Master’s thesis applied the ERASOR to 78 adolescent males. Although the total score was not found to be predictive of sexual recidivism (AUC=.59), it did significantly predict violent (including sexual) reoffending (AUC=.65) (Morton, 2003).
•A doctoral dissertation by Rojas Mejia (2013) applied the ERASOR to 100 males. The total score was predictive of violent (sexual and non sexual) recidivism with an AUC of .67. Adolescents rated as high risk reoffended with a sexual offence at a faster rate than those rated as low risk.
•An unpublished thesis applied the ERASOR to 93 adolescent males, yielding AUCs of .48 and .49 for the total score and clinical judgment ratings respectively (Nelson, 2011).
Validation History
•Skowron (2004) tested the ERASOR on 110 adolescent males. It significantly predicted any reoffending (AUC=.67), any nonsexual violent offence (AUC=.68) and any sexual recidivism (AUC=.71).
Applicability: Females
•A doctoral dissertation found the ERASOR total score was not predictive of sexual recidivism (AUC=.50) when applied to 128 adolescent males (McCoy, 2007).
•Worling and Langton (2015) evaluated scores from a sample of 81 adolescent males with at least one sexual offence. Findings showed the ERASOR was significantly correlated with sexual recidivism in a follow up period of on average 3.5 years.
•The ERASOR is currently in widespread use throughout Canada and the United States and a number of other countries.
•Multiple studies have been carried out on the ERASOR by authors other than the tool developers. Mixed findings in previous validation studies regarding its predictive accuracy; although more studies demonstrated good than poor results
•Edwards et al. (2012) found that the ERASOR can be useful in monitoring treatment progress, with significant differences in ERASOR total scores between those who do or do not reoffend.
•Can be time consuming to complete.
•The tool can help assessors develop offence analyses and risk management plans.
Validation History
a)UK Research
Contribution to Risk Practice
No empirical evidence at present.
b)International Research
Other Considerations
a)UK Research
RATED page updated: August 2019
No empirical evidence at present.
•For more information regarding the ERASOR, supporting documents and for updated research support, please visit radiuschild youthservices.ca. Electronic copies of the ERASOR can also be accessed for free via the website and contact can also be made regarding the tool at this site.
© Risk Management Authority 2019
•The ERASOR can help determine the level of monitoring/rehabilitative efforts required to manage the risk posed by the individual.
Applicability: Mental Disorders
No empirical evidence at present.
b) International Research
•Chu et al. (2011) in a sample of individuals from Singapore, the ERASOR composite score achieved moderate to high predictive accuracy in relation to sexual (AUC = .74) and non sexual (AUC = .66) recidivism. The ERASOR clinical ratings obtained AUC values of .83 and .69 for sexual and non sexual recidivism respectively.
No empirical evidence at present.
Applicability: Ethnic Minorities
Validation History
b)International Research
•The ERASOR can aid assessors in identifying risk factors. Some of the factors included in the ERASOR can act as targets for change. These factors can also contribute towards the measurement of progress or deterioration in factors related to the individual’s level of risk.
For boys aged 12 18
•Items are scored on a 3 point Likert scale of 0, 1 and 2 depending on the extent to which the factor is present.
Year 2003
•Measures dynamic variables as well as static ones, which allows for the assessment of change (i.e. progress in treatment) and also informs intervention needs and targets (Yates, 2005).
•The items are grouped under four scales: (1) Sexual Drive/Sexual Preoccupation, (2) Impulsive/Antisocial Behaviour, (3) Clinical/Treatment and (4) Community Adjustment.
Description
•The J SOAP II is a 28 item checklist of risk factors designed to assess risk of sexual violence and general delinquency in male adolescents with a history of sexually coercive behaviour and/or convictions for sexual offences.
Name of Tool Juvenile Sex Offender Assessment Protocol II (J SOAP II)
RATED page updated: August 2019 © Risk Management Authority 2019
Author / Publisher Prentky and Righthand
Category Youth Assessment: Sexual Violence Risk (Validated)
Assessors must possess the relevant training/experience in youth assessment pertaining to sexual offending, in particularly adolescent development, risk assessment and assessing juveniles with sexual offending. In the manual, it is recommended that assessors liaise with each other intermittently to discuss scoring and keep themselves informed about the recent literature pertaining to juvenile sexual offending.
Strengths
•The J SOAP II total and subscale scores can be reported as ratios or proportions reflecting the observed amount of risk rated at a given point in time. The J SOAP II does not contain cut off scores or generate estimates of probability (Viljoen et al., 2017).
Age Appropriateness
Assessor Qualifications
•The risk assessment variables were developed from research reviews of literature covering 5 areas: (1) clinical studies of juvenile who had sexually offended, 2) risk assessment/outcome studies of juveniles who had sexually offended, 3) risk assessment/outcome studies of adults who had sexually offended, 4) risk assessment/outcome studies from the general juvenile delinquency
Empirical Grounding
b)International Research
Inter Rater Reliability
General Predictive Accuracy
literature, 5) risk assessment studies on mixed populations of adults who have offended (Prentky and Righthand, 2003: 2).
None available at present.
RATED page updated: August 2019 © Risk Management Authority 2019
•Aebi et al. (2011) found good inter rater reliability for the total index score (ICC = .71)
Validation History
a)UK Research
a)UK Research
•Viljoen et al. (2017) assessed the inter rater reliability of the Intervention and Community Stability/Adjustment scales in the J SOAP II. A sample of thirty seven adolescents yielded an ICC range of .64 to .82, showing good to excellent inter rater reliability.
•Barroso and colleagues (2019) examined the inter rater reliability of the Portuguese version of the J SOAP II and found that this was good to excellent ranging from .73 to .81.
None available at present.
•Martinez, Flores and Rosenfeld (2007) the J SOAP II composite score demonstrated good inter rater reliability (ICC = .70).
•In a study of 166 juveniles who were followed up over an average time period of 10.75 years, Schwartz Mette and colleagues (2019) assessed inter rater reliability using a subset of the sample (n=36). Moderate to good IRR was evident for each of the components: Scale 1 (Sexual Drive/Preoccupation), ICC=.78; Scale 2 (Antisocial Behavior/Impulsivity), ICC=.90; Scale 3 (intervention), ICC=.64; Scale 4 (Community Stability/Adjustment), ICC=.54, Static scale, ICC=.90, Dynamic scale, ICC=.58 and Total Score, ICC=.78.
•Chu et al. (2012) obtained an excellent ICC of .77 for the composite J SOAP II score.
•Wijetunga et al. (2018a) found there was good IRR when using the J SOAP II, with a total scale of .88 and a static summary range of .74 .90.
•Maximum likelihood logistic regression analyses were conducted by Viljoen and colleagues (2017) to test the outcome of any reoffending. A lack of relationship between changes scores in the J SOAP II and reoffending rates led them to conclude that the J SOAP II may not adequately capture the relevant dynamic factors. When risk factors decreased, however, the J SOAP II Intervention scale was found to significantly predict lower rates of sexual reoffending (OR=0.14, p=.013).
•Viljoen, Mordell and Beneteau (2012) in a meta analysis, the J SOAP II composite score attained moderate AUC values of .67 for sexual reoffending and .66 for general reoffending respectively.
b)International Research
RATED page updated: August 2019 © Risk Management Authority 2019
•Viljoen and colleagues (2008) found that the J SOAP II was less significant in predicting re offending among younger adolescents. Adolescents aged 15 and younger were more likely than older adolescents to be incorrectly identified as being high risk for sexual and nonsexual violence following discharge.
•A study in Singapore concluded that the J SOAP II only had limited utility for predicting sexual recidivism in a non Western context: the Sexual Drive/Preoccupation scale was the only significant indicator. Conversely, it did appear to have significant predictive validity for assessing non sexual recidivism (Chu et al., 2012).
•Aebi et al. (2011) found that total score showed moderate predictive accuracy for sexual recidivism (AUC=.645) and only a small effect for nonsexual violence and general recidivism (AUCs of .633 and .607 respectively). Further to this, ROC analyses revealed that sexual recidivism was significantly predicted by the J SOAP II antisocial, adjustment and Sexual Offence Severity (SOS) scales with AUCs of .739, .743 and .751 respectively; however, this did not extend to the remainder of the J SOAP II scores or the number of sexual assaults against the index victim(s).
•In a comparative study between a medium security correctional setting and an unlocked residential sexual offending treatment programme, it was determined that there were no significant differences between the sites. The overall predictive accuracy of post release sexual offending arrests was found to be modest with an AUC of .64 (Martinez et al., 2015).
•Wijetunga et al. (2018b) compared and contrasted the predictive validity of the J SOAP II based on age groups and sex drive levels (as measured by item 7 on the J SOAP II). It was found that the tool was an adequate predictor of sexual recidivism for younger juveniles (14 16 years) than older ones (17 years and older). In terms of sex drive, adequate predictive accuracy was found for those with a heightened sex drive (AUC=.70); although the predictive accuracy was poor for those with a lower one (AUC=.58).
•Wijetunga et al. (2018a) created a psychopathy scale (Scale P) (intended to assess psychopathy) and included it in their study of the J SOAP II to test predictive accuracy of this combined measure in 72 juveniles with sexual offences. The scale is not part of the J SOAP II and includes items that assess psychopathy. For general nonsexual, violent nonsexual and sexual recidivism, AUCs of .75, .69 and .73 were generated. These were significantly higher that the AUCs for the J SOAP II total score, which were .61, .53 and .72 for general nonsexual, violent nonsexual and sexual recidivism respectively. This suggests that inclusion of items that assess psychopathy may enhance the clinical utility of the J SOAP II. Further research is required, however, to properly validate this finding; particularly given the small sample size in this study.
•Viljoen et al. (2008) the J SOAP II demonstrated poor to moderate accuracy in predicting recidivism with AUC values ranging between .46 to .58.
•The J SOAP II was tested for 166 juveniles over a 10.75 year period, following them into adulthood. The J SOAP II
•Rajlic and Gretton (2010) found that the J SOAP II composite score has moderate to high predictive accuracy in relation to sexual (AUC = .69) and non sexual (AUC =.77) recidivism. It also found that the J SOAP II was not predictive for youth with both sexual and nonsexual offences, suggesting that typological differences may exist.
•Prentky et al. (2010) in a 7 year follow up, the authors compared and contrasted two higher risk subsamples of pre adolescents (aged 11 years and under) and adolescents (aged 12 years and over). The composite score generated large predictive accuracy with AUCs of .80 and .83 for pre adolescents and adolescents respectively.
RATED page updated: August 2019 © Risk Management Authority 2019
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
Validation History
Applicability: Mental Disorders
Not intended for use with females.
Applicability: Females
Validation History
b)International Research
None available at present.
No empirical evidence available.
a)UK Research
Applicability: Ethnic Minorities
Contribution to Risk Practice
•Martinez, Flores and Rosenfeld (2007) found predictive accuracy between the composite score and ‘any’ re offence (AUC = .76) and sexual re offence (AUC = .78) in a sample of individuals of African American (63.5%) and Latino (14.7%) ethnic origin. The remainder of the sample were Caucasian (14.7%) or other/unknown (1.9%).
Total Score, Scale 1 and Static Score were each significantly associated with new sexual charges (AUCs of .76, .77 and .79). Non significant results emerged from the rest of the scales. For nonsexual, violent reoffending, all scales bar Scale 1 were significant: Total Score, AUC=.68; Scale 2, AUC=.68; Scale 3, AUC=.66; Scale 4, AUC=.66. With regards to any other offending (nonsexual and nonviolent), Scales 1, 4 and the Total and Static Scores did not demonstrate predictive validity. Scales 2 and 3 and Dynamic Scores yielded AUCs of .63, .60 and .60 respectively (Schwartz Mette et al., 2019).
•Chu et al. (2012) in a sample of individuals from Singapore, the J SOAP II total score had good predictive accuracy in relation to non sexual recidivism (AUC =.79); however, it was unable to significantly predict sexual recidivism.
Other Considerations
•The findings of the study by Barroso et al. (2019) indicate that the J SOAP II can be adapted to different languages. A Portuguese version of the instrument was found to be conceptually equivalent, show acceptable psychometric properties and perform similarly.
•For more information on the J SOAP II please contact the authors, Dr. Robert Prentky or Dr. Sue Righthand.
•The J SOAP II can aid assessors in identifying risk and responsivity factors specific to the individual (e.g. ‘motivation to change’).
•The tool may be useful in informing treatment and/or interventions and guiding risk management decisions.
•The tool’s dynamic scales can help to measure the individual’s progress through treatment.
RATED page updated: August 2019 © Risk Management Authority 2019
•Some of the factors included in the tool can act as targets for change.
•No cut off scores have been generated for the J SOAP II authors recommend that judgments of the youths’ risk of re offending not be made exclusively on the basis of their J SOAP II scores (Righthand et al., 2005). Cut off scores may also be misleading as they do not take into account false positives and false negatives (Righthand, personal communication, January 2013).
•The J SOAP II is aimed at facilitating short term case management and intervention goals, so it may be limited in informing long term decisions (Prentky et al., 2010). Ralston and Epperson (2013) highlight the difficulty in making longer term predictions on the basis of adolescent behaviour, by testing both adult and juvenile sexual offending tools, including the J SOAP II, on juveniles who sexually offended. The accuracy of longer term predictions of adult sexual recidivism was substantially lower than that achieving in predicting the sexual recidivism of juveniles.
•Scales 2 (related to general delinquency) and 3 (associated with treatment and progress, e.g. accepting responsibility) of the J SOAP II were found to have concurrent validity with other youth instruments, the PCL:YV and the YLS/CMI (Barroso et al., 2019).
•The J SOAP II manual cautions that decisions regarding an individual’s risk of reoffending should not be based solely on the results generated by the tool. The J SOAP II should instead be used as part of a more comprehensive risk assessment process.
•The J SOAP II can contribute to risk management measures such as victim safety planning and contingency planning.
•The developers of the tool maintained that it is an empirically informed guide to facilitate the systematic review and assessment of items that may predict an increased risk of reoffending and assist with choosing treatment options. They caution that the J SOAP II is not to be used as an actuarial scale and it does not provide cut off scores for categories of risk.
MEGA♪ can be completed by licensed mental health professionals or non clinical professionals (e.g., child welfare workers, probation officers, residential support workers). Assessors must have at least 2 years of experience working with sexually abusive youth prior to using the tool, and must complete a one day certification training.
Year 2006 Description
RATED page updated: August 2019 © Risk Management Authority 2019
Name of Tool Multiplex Empirically Guided Inventory of Ecological Aggregates for Assessing Sexually Abusive Adolescents and Children (MEGA♪)
Author / Publisher Miccio Fonseca, L. C.
•MEGA♪ risk assessment tool is the first to simultaneously assess risk levels for coarse sexual improprieties (i.e. sexually vulgar comments, expressions and behaviours) and/or sexually abusive behaviours and protective factors in youth. It is an outcome measure assessing a youth’s progress, with re assessments taking place every 6 months to compare changes in the youth’s risk levels and protective factors.
•MEGA♪ generates a computerized scored comprehensive risk assessment report idiosyncratic to the youth assessed, a feature not seen in other risk assessment tools. The reports are appropriate for use in forensic settings to provide information to the court related to baseline risk level and changes in risk and protective factors over time.
Category Youth Assessment: Sexual Violence Risk (Validated)
•There are two types of reports. MEGA♪ Individualized Risk Assessment Report identifies the baseline risk level specific to the individual’s risk of engaging in sexually abusive behaviours and protective factors that mitigate risk. The MEGA♪ Individualized Outcome Risk Assessment Report provides a comparative analysis of changes in baseline risk level and protective factors over the last 6 months.
•MEGA♪ incorporates inquiry relating to questionable, sexually related internet activities, such as sexting and revenge porn, and/or posting inappropriate sexual content on social media (Miccio Fonseca, 2017b, 2017d).
•MEGA♪ established four levels of risk. ‘Very High’ risk has a number of substantially persistent and concerning variables present for potential risk for coarse sexual improprieties and/or sexually abusive behaviours, likely at very critical behaviours requiring immediate intervention. For instance, sexual violence including physical threats and bodily harm, use of a weapon and luring, stalking and/or torturing victims would fall into the ‘Very High Risk’ category (Miccio Fonseca, 2017d). The ‘Very High’ risk level is designed to differentiate youth who are sexually violent and/or predatory violent, including those who are sex traffickers (male or female) (Miccio Fonseca, 2017c, 2017e).
Age Appropriateness 4 19.99 years
Assessor Qualifications
•MEGA♪ caters to all levels of developmental and cognitive ability. It is applicable to youth ages 4 19 years, adjudicated or non adjudicated (males, females, and transgender females, including youth with low level of intellectual functioning) (Miccio Fonseca, 2009, 2010, 2013, 2016a, 2016b, 2017a, 2017b, 2018a, 2018b, 2018c, 2019).
•Applicable to males, females, and transgender females.
RATED page updated: August 2019 © Risk Management Authority 2019
Empirical Grounding
•The tool is able to track youth over time and do comparative analysis on changes in risk level and protective factors every 6 months.
•Applicable to youth with low intellectual functioning
In each research site, each item was collectively reviewed by professionals to accommodate cultural nuances in language and clarify differences in terms (e.g. educational levels, ethnicity classification, type of adjudication, probation and type of weapons) (Miccio Fonseca, 2013).
•MEGA♪ established normative data and calibrated risk levels grounded on given algorithms according to age and gender; no “guess estimates” on the youth’s level risk. The risk level assessed is definitive.
Case vignettes were used internationally to analyse the inter rater reliability of the MEGA♪ within the cross validation study. Scoring of the assessment achieved 98% 100% agreement by those who received MEGA♪ training (Miccio Fonseca, personal communication, January 2013).
a)UK Research
•Applicable to pre adolescents (youth under 12 years).
•Applicable to adjudicated (those on whom a formal legal decision has been made) and non adjudicated youth
Case vignettes were used to analyse the inter rater reliability of the MEGA♪ in the sites in England and Scotland. The tool achieved 98% 100% agreement in the scoring of the assessment by those who received MEGA♪ training on the tool (Miccio Fonseca, personal communication, January 2013).
Strengths
•All cross validation studies demonstrated significance on predictive validity.
The Fonseca Inventory of Sex Offender Risk Factors (FISORF 1998; Miccio Fonseca, 2005) provided the blueprint of the ecological framework design for MEGA♪ The empirically guided variables for the FISORF 1998 and MEGA♪ came from two sources (a) extensive quantitative review of the literature; (b)qualitative clinical interview data from a 7 year (1988 1995) descriptive research study of youth and adult, ages 4 72 (n=656; 72% of the sample under age 18) (Miccio Fonseca, 1996). The selected MEGA♪ items were compared against ‘best marker’ variables identified in logistic regression analysis of the JSORRAT II construction sample (Epperson et al., 2006; Epperson and Ralston, 2015) Construct validity with JSORRAT II was established in the MEGA♪ validation study (Miccio Fonseca, 2009, 2010).
•Tested and retested on large ethnically diverse representative samples (over 4,000 youth); making the findings generalisable.
•Can be used in forensic settings.
•MEGA♪ simultaneously assesses risk levels and protective factors.
b)International Research
Inter Rater Reliability
a)UK Research
General Predictive Accuracy
b)International Research
RATED page updated: August 2019 © Risk Management Authority 2019
•A cross validation study by Miccio Fonseca (2013) was carried out in several sites in the UK including Scotland and England (further information is available below).
Validation History
•In the major study sample of 2717, 12 transgender female youth were present. appear to have more varied sexual experiences and contact than their male/female counterparts. Moreover, there were a greater number of incidents involving adults or adults and children (transgender females 17%, males 4% and females 3%), as well as those involving more than two victims (transgender females 50%, males 27% and females 10%). The results indicate that approaches to sexually abusive transgender female youth should be tailored accordingly (Miccio Fonseca, 2018b).
Validation History
•An independent 6 year longitudinal study by Rasmussen (2017) compared MEGA♪ and J SORRAT II. The study was on adjudicated male adolescents (n= 129) in an intensive facility for sexually abusive youth. The study indicated both tools had predictive validity; AUC for the MEGA♪ was .67; whilst the JSORRAT II was not predictive.
Applicability: Females
None available at present.
•Three cross validation studies (Miccio Fonseca, 2016b, 2017a, 2018a), as well as two major studies with combined samples (Miccio Fonseca, 2017a, 2017b, 2017d, 2018c) were carried out. In all five studies, MEGA♪ consistently demonstrated consistent predictive accuracy (AUCs ranging from .71 to .91).
a)UK Research
•A cross validation study was conducted on 1,056 young persons (males and females), 238 of whom were identified through information in the case file as having low intellectual functioning. Sample consisted of youth from 13 research sites from several different countries, which included the US, Canada, (and the UK). The Risk Scale obtained moderate accuracy in predicting sexual recidivism in the age groups 4 12 (n=39) and 13 19 (n=334) years old (AUC of .77 and .71 respectively). It was also found that risk levels increased with age, with those aged 13 15 scoring higher than youth aged 4 12 years old (Miccio Fonseca, 2013).
a)UK Research
None available at present.
b)International Research
•Studies have included males, females, and transgender females (Miccio Fonseca 2009, 2010, 2013, 2016a, 2017a, 2017b)
RATED page updated: August 2019 © Risk Management Authority 2019
•The MEGA♪ was validated and cross validated on large samples of youth (N=1184 and 1056 respectively) aged between 4 and 19. The samples also included approximately 20% of youth with low intellectual functioning (Miccio Fonseca, 2009, 2010, 2013).
Contribution to Risk Practice
Validation History
Applicability: Mental Disorders
•Miccio Fonseca (2016a) reported females were found to present more psychological difficulties in terms of higher incidences of depression and negative affect in the last six months.
a)UK Research
Applicability: Ethnic Minorities
None available at present.
b)International Research
•An independent study by Fagundes (2013) examined the association between the risk levels from MEGA♪, J SORRAT II and DSM IV diagnosis. Findings were that there was a significant correlation between the two tool’s ability to measure the same aspect of risk (r = 0.48).
•Gender comparisons showed that males scored higher than females. Moreover, there were also differences between age groups: for females, it was found that the older the youth, the lower the protective score (Miccio Fonseca, 2009, 2010).
Validation History
b)International Research
•The MEGA♪ has been tested and retested on large ethnically diverse representative samples (over 4000 youth from USA and several countries); all cross validation studies demonstrated prognostic utility making the findings generalisable (Miccio Fonseca, 2009, 2010, 2013, 2016a, 2016b, 2017a, 2017b, 2017d, 2018a)
•The MEGA♪ can be completed by licensed mental health professionals or non clinical professionals (e.g., child welfare workers, probation officers, residential support workers). However, assessors must have at least 2 years of experience working with sexually abusive youth prior to using the tool.
•MEGA♪ has seven aggregates related to risk for coarse sexual improprieties and/or sexually abusive behaviors, each providing an accumulation of information on particular targeted areas in need of attention for the youth being assessed (i.e., Neuropsychological, Family Lovemap, Antisocial, Sexual Incident, Coercion, Stratagem, and Relationship [Predatory Elements]). Items are rated as either yes or no.
•For further information contact L.C. Miccio Fonseca, Ph.D. via email: lcmf@cox.net
•The MEGA♪ Specialized Risk Assessment, 1 Day Certification Training is required to use the tool.
RATED page updated: August 2019 © Risk Management Authority 2019
Other Considerations
•MEGA♪ empirically establishes a fourth level of risk, for those sexually abusive youth who are the anomalies (i.e. most dangerous and potentially lethal, sexually violent and/or predatory sexually violent) (Miccio Fonseca and Rasmussen, 2009, 2014, 2018).
•MEGA♪ is composed of four distinct scales (reflecting the incorporated seven aggregates): (a) Risk Scale, (b) Protective Scale, (c) Estrangement Scale, and (d) Historic Correlative Scale (formally Persistent Sexual Deviancy Scale).
There is a competency requirement for using the AIM Under 12s Model; only those approved by The AIM Project can use the Model. The rationale behind this is to provide a quality assurance for commissioners, increase the confidence of practitioners and ensure the quality of assessments undertaken. Potential assessors must undertake the relevant training run by The AIM Project and pass the competency requirement in the training in order to be approved by The AIM Project to use the models. It is expected that those attending the training should have relevant practice experience of complex assessments, ideally of sexual behaviour (Carson, 2019).
•Pattern Mapping is a visual framework capturing key life events and sexual behaviours
•The AIM Under 12s model for children aged 8 12 years old, is a framework for professional analysis and decision making. There are five Domains, with five Factors within each Domain looking at different aspects which need to be considered to give an overall outcome for that Domain. In addition, each Factor has a number of items provided to support the professional analysis, and these can be added to by information which is unique to a particular child or their family. Within every Domain professionals are asked to consider both strengths and concerns (Carson, 2019).
8 12 AIM Under 12s Assessment Model
The AIM Project Assessment Models for Children under the age of 12 years old (3rd edition)
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
Author / Publisher Carol Carson/The AIM Project
•The factors are scored as follows: zero, no general concern or it is an area of strength; two, for some concern; four, where there is significant concern. The scores for all 5 Factors within each Domain are then collated to give an outcome for the Domain. Once all 5 Domains are scored, this provides a visual profile graph of the child in their context indicating areas of significant concern which would be red (scores of 14 20); areas which indicate work is required which would be amber (scores of 6 12) and areas which are not a concern or are potentially strengths which would be green (scores of 0 4) (Carson, 2019).
•The AIM Project offers two models for children under 12 years old: one for children under 7 years (Pattern Mapping), the other for older children aged 8 12 years old, which is a dynamic risk assessment model framework not an actuarial risk assessment tool Both models are designed to be visual and easily updated at review points to highlight progress being made to facilitate communication with parents, children and professionals. (Carson, 2019)
Year 2019 Description
Assessor Qualifications
Name of Tool
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
0 7 years Pattern Mapping
•The dynamic nature of the AIM Under 12s means that historical information is assessed in relation to how it impacts upon current concerns. The purpose of this is to avoid children being ‘stuck in time’ since the child may have developed, matured or changed since the time of the historical information (Carson, 2019).
•Pattern mapping is advised for younger children under 7 years old. This is a visual framework inquiring about the who, what, where, why and when of sexual behaviours to facilitate professionals in understanding the following:
RATED page updated: August 2019 © Risk Management Authority 2019
General Notes
•The AIM Under 12s stages are: immediate risk management safety plan, establishing rapport with parents/carer and the child, pattern mapping with the professional group; interviewing the parent/carer and interviewing the child (Carson, 2019).
5)Self regulation which addresses the factors the child inherently has or could develop which may help them to manage or stop their sexual behaviours (Carson, 2019).
Causal factors for the sexual behaviour how and where the child may have learnt the behaviour.
Meaning of the behaviour to the child do they understand their behaviour is sexual? Is it more about emotional needs; attachment; belonging etc.
3)Developmental, looking at the child’s pathway into the sexual behaviour considering their own childhood issues such as possible abuse or adverse childhood experiences and their families
•The AIM Under 12s model is recommended for children aged 8 12 years, it shares the same AIM framework of Domains and Factors as AIM3 but is based on research and practice with children and takes into account their developmental level. This consists of Five Domains:
2)Non sexual behaviour, looking at antisocial behaviour, mental health and general behaviour;
Tool Development
4)Environmental/family,functioning
Patterns to the behaviour who is at risk; when and where does the behaviour happen; what patterns are emerging as to how it happens, triggers for the behaviour and gaps when the behaviour does not happen.
•Pattern mapping and the AIM Under 12s model may be used with both boys and girls, with the language used in the book being gender neutral. The AIM Under 12s may also be used with children
considering the ability and willingness of the adults around the child to provide safety and security and to both support and supervise the child to help them change or cease the sexual behaviour. It also considers the other key people around the child and adults who may support or hinder any Safety Plans or interventions.;
Motivation are they motivated to engage with adults regarding their behaviours? This technique allows for the tracking patterns and meanings across the life course of a young child, looking at whether life events and sexual behaviours are linked (Carson, 2019).
•The AIM Under 12s was first published in 2007, with subsequent updates following in 2014 and 2019. The AIM Under 12s was designed so it allows for progress and change to be visually represented to the young person, their family and professionals. The checklist of ‘normal,’ ‘appropriate,’ ‘inappropriate’ and ‘harmful’ is based on Hackett’s (2010) ‘Continuum of Sexual Behaviours. It recognises that children should not be assessed solely on their sexual behaviours (Carson, 2019).
1)Sexual Behaviour, looking at the nature and extent of the sexual behaviour including characteristics of victims;
•For further information about the AIM Under 12s, visit http://aimproject.org.uk/ or email admin@aimproject.org.uk
RATED page updated: August 2019 © Risk Management Authority 2019
with learning disabilities and those on the autistic spectrum with caution, given the research and practice experience on which the model is based relates to mainstream children (Carson, 2019).
RATED page updated: August 2019 © Risk Management Authority 2019
•AIM3 is a 25 item assessment framework designed to help practitioners consider relevant targets for intervention, in addition to quantifying risk and levels of supervision. It is not an actuarial risk assessment tool (Leonard and Hackett, 2019).
•Foregoing the previous risk levels of low, medium of high, the scoring of factors as 0, 2 or 4 acts as a guide for the assessor. Each domain has a maximum of 20 points available, with scoring ranging from 0 to 100. Totalling up scores across domains is categorised in a colour matrix: red, scores of 14 20, which may indicate an area of relative need or risk requiring specific or immediate intervention; amber, scores of 6 12, which may indicate the need to lower risk and meet needs requiring intervention in the medium term; green, score of 0 4, which may indicate an area of relative strength in the individual’s presentation/context, something which may be utilised to support interventions with the individual. The assessor is to use their professional judgment to reach a final decision about the individual’s risk level, with the use of AIM3 having the potential to act as a guide for risk management, interventions and safety planning (Leonard and Hackett, 2019).
Age Appropriateness
•Whilst acknowledging the importance of historical information, AIM3 looks at the impact of historical factors on the current presentation and functioning of the individual being assessed. This allows for a more dynamic assessment, whereby historical factors are considered for their relevance to the individual at the present time (Leonard and Hackett, 2019).
•Items are organised into five domains: sexual behaviours; non sexual behaviours; developmental factors; environmental/family influences; self regulation. Use of the AIM3 helps to develop an overall profile of a young person across the five domains (Leonard and Hackett, 2019).
Author / Publisher Leonard and Hackett/The AIM Project Year 2019
Assessor Qualifications
Young people aged 12 18
•The unique characteristics of victims (e.g. race, gender, learning disabilities) should be considered when using the AIM3 (Leonard and Hackett, 2019).
Name of Tool AIM3
•The instrument is appropriate for use with young males aged between 12 and 18 years old who are known to sexually abuse. This includes contact and technology assisted sexual offences (e.g. downloading indecent images of children). It may also be used with young women, with a degree of caution (Leonard and Hackett, 2019).
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
Description
Potential assessors must undertake the relevant training run by The AIM Project and pass a competency requirement, in order to be approved by The AIM Project to use the Assessment Model. Training involves a competency case study and includes copies of the book. It is expected that those
•AIM3 has superseded AIM2 in practice. Until AIM3 training is undertaken, AIM2 may be used in the interim, but it will become out of date.
The competency requirement for using the AIM3 is to provide a quality assurance for commissioners, increase the confidence of practitioners and ensure the quality of assessments undertaken.
(2) Non Sexual Behaviours, measuring the following: general criminality that is non sexual in nature; non sexual aggression and antisocial behaviour; alcohol and drug uses; general behaviour as well as mental health and wellbeing;
•The AIM3 can be used to assess young people in custody of secure care environments; although, caution should be taken when scoring the factors taking the context of the individual’s environment into account. It is also recommended that the AIM3 is reviewed prior to release (Leonard and Hackett, 2019).
RATED page updated: August 2019 © Risk Management Authority 2019
Tool Development
attending the training should have relevant practice experience of complex assessments, ideally of sexual behaviour
•The AIM3 has been piloted successfully and has worked well in practice. A research paper is in development for this (AIM Project 2019, personal communication).
(4) Environmental/family, examining the effect of the environment and wider social and family context in which they live: stability and safety; parental or carer supervision; relationships and peer groups; education, employment and leisure.
(5) Self Regulation, detailing how the individual functions in terms of their abilities to understand the impact of their behaviour and their self regulation skills: responsibility; motivation and engagement; future perspective; problem solving; social competence (Leonard and Hackett, 2019).
General Notes
•As AIM3 was just launched in July 2019, there is no current research relating to the AIM3.
•The AIM3 covers five domains:
(1) Sexual Behaviours (offence specific), looking at: the nature and extent of this behaviour; the characteristics of victims; sexual aggression; the range of sexual knowledge, attitudes and interests.
•The original AIM was published in 2000. The assessment was based on the Risk Aetiology Model (Beech and Ward, 2004). A second version to the AIM was published in 2007, following various refinements made to the original AIM assessment It was further refined in 2012 to make it relevant to females and those with a learning disability. This further revision in 2019, draws upon feedback from practitioners and in response to changes in practice, e.g. the increasing use of technology in everyday life as well as potentially within harmful sexual behaviour. This led to the removal of the question mark function, the rating of low, medium or high and allows practitioners to consider responsivity issues. The revision also allowed for the incorporation of new research and to allow the assessment to be more fluid and adaptable in line with an individual’s progress.
(3) Developmental, looking at influences on these wide ranging behaviours: trauma and victimisation; childhood and adolescent adversity; attachment; family functioning; health, intellectual and emotional functioning.
•It is recommended that the AIM3 is used to review the progress a young person makes over time, with the intention to reduce the domains to ideally ‘green level’ but at the very least the ‘amber level.’ The AIM3 can be utilised in supervision to aid individualised interventions for the individual and their family, allowing the young person to visualise their progress and plan the next steps.
•The authors of AIM3 advise that it can be used, with caution, with young women. The practitioner should give consideration to the behaviours and attitudes that a young woman may present that could differ from that of a male. Similarly, when using the AIM with Black Asian and Minority Ethnic groups (BAME), practitioners should be mindful of the cultural and religious practices relevant to the individual and consider the impact of this on the young person’s sexual and non sexual development, family structure, environment and self regulation (Leonard and Hackett, 2019).
•For enquiries regarding the AIM3 manual, contact admin@aimproject.org.uk
•For publication updates, please visit www.aimproject.org.uk
•The AIM3 can be used for technology assisted harmful sexual behaviour (TA HSB) internet offences where there are also instances of direct or non contact harmful sexual behaviour. To clarify, where there is ‘DUAL HSB’ referring to young people engaging in harmful sexual behaviour and who also use technology to assist their HSB. In these types of assessments, thus should be used in conjunction with the TA HSB framework, a case formulation model (Allotey and Swann, 2019). In incidences where there are only internet offences, the TA HSB guidance should be used. It is strongly recommended that practitioners also undertake the training offered in TA HSB (Leonard and Hackett, 2019).
•The AIM Project divides its courses into Foundation and Advanced levels to allow practitioners to attend at the appropriate level for their knowledge and expertise. The AIM3 is an Advanced level course.
•AIM3 can help guide interventions. The AIM Project has also provided guidance on interventions, setting out a framework of four stages for Interventions and Safety Plans (Guilnermino and McCarlie, 2019). These may be used for different groups: for instance, sibling abuse, intergenerational abuse (i.e. incest). It is recommended that interventions are holistic in nature, being informed by factors associated with resilience and positive outcomes.
•In addition to the UK and Ireland, AIM3 has been implemented in New Zealand and Norway, with interest from Canada, Australia, Spain and Germany (Carol Carson 2019, personal communication).
•The AIM3 can be used with young people with learning disabilities with strong caveats (e.g. adapting interview style and language used). The learning disability must be considered as part of the analysis. Similar conditions apply with using the AIM3 with a young person with autism (Leonard and Hackett, 2019).
RATED page updated: August 2019 © Risk Management Authority 2019
•Practitioners should gain as much information as possible about the harmful sexual behaviour, accessing victim statements/accounts where possible. It is essential that practitioners hold knowledge of the criteria for different sexual offences. For instance, harmful sexual behaviour involving coercive and non consensual penetration or attempted penetration may use higher levels of psychical force or emotional manipulation against their victims.
•In an earlier study, eight factors were found to differentiate between children who did and did not continue engaging in concerning sexual behaviours following adult reprimand. A total score based on the eight factors attained an AUC value of .86 in predicting those who continued engaging in concerning sexual behaviours (Curwen and Costin, 2007; Curwen, Jenkins and Worling, 2009). The author advises that these eight factors alone should not be used to determine level of concern (risk)
Age Appropriateness
•The AR RSBP is a 34 item risk assessment tool designed to assist in the assessment of the level of concern (risk) that a child will continue engaging in sexual behaviour problems (SBP).
Under 12 years (not appropriate for 12 year olds).
•The static and dynamic AR RSBP items are organized into 6 categories: 1) Sexual Behaviour Characteristics, 2) Victimization Experiences, 3) Violence and Control, 4) Personal and Interpersonal Characteristics, 5) Family Characteristics, and 6) Intervention.
•The first version of the tool (RSBP) was completed in 2006. Since its inception, the AR RSBP has undergone revisions to its structure and content (Curwen, 2011a).
Assessors should have good knowledge of child development and child sexual development.
Tool Development
•Empirical and clinical evidence specific to: 1) children known to have continued SBP after identification; 2) children believed likely to continue; 3) treatment goals for children with SBP; 4) assessment tools designed for children at risk for multiple behaviours that included sexual; 5) research regarding adolescent with sexual offences who commenced SBP during childhood.
•Curwen (2011b) found that children who had been reprimanded and then repeated concerning sexual behaviours had significantly higher RSBP total scores than children who had not repeated these sexual behaviours subsequent to reprimand.
Assessor Qualifications
Name of Tool Assessing Risk to Repeat Sexual Behaviour Problems Version 2.1 (AR RSBP)
•The AR RSBP is intended for use with male and female children under the age of 12
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
•The tool was modelled on the ERASOR and is based on empirical evidence and professional opinion; there are no cut off scores or formulas for determining level of concern (risk).
RATED page updated: August 2019 © Risk Management Authority 2019
Author / Publisher Curwen Year 2010 Description
•The tool was designed to be used as part of a comprehensive assessment.
•Radius Child and Youth Services is conducting research on the reliability and validity of the RSBP. To learn more about Radius’ AR RSBP research contact tcurwen@radiuschild youthservices.ca
RATED page updated: August 2019 © Risk Management Authority 2019
•For further information, please visit the following website: https://faculty.nipissingu.ca/t/) or contact Dr. Tracey Curwen (tcurwen@nipissingu.ca)
General Notes
•Hempel et al. (2013) the JRAS did not significantly predict sexual recidivism.
•Designed to assess sexually abusive youth who are adolescents
RATED page updated: August 2019 © Risk Management Authority 2019
Tool Development
•Caldwell, Ziemke and Vitacco (2008) the JRAS demonstrated excellent inter rater reliability (ICC) of .94. Despite this, the tool was unable to predict sexual, non sexual, violent or general offending.
•Hiscox, Witt and Haran (2007) the JRAS had moderate inter rater reliability (r =.66). It also demonstrated small correlations with sexual recidivism (r =.15). The sexual deviance factor did not predict recidivism (both sexual and no sexual); the major predictive factor in the JRAS was the antisocial behaviour one.
Name of Tool
Assessors should undertake the appropriate training prior to administration of the tool.
Description
12 to 19
Year 2010
•The JRAS is a 14 item scale designed to assess the risk of reoffending sexually among males who have been adjudicated for a sexual offense It is used by New Jersey to place sexually abusive youth into risk tiers in accord with Megan’s Law.
Assessor Qualifications
•Risk is characterised as ‘low,’ ‘medium’ and ‘high.’
•Ralston and Epperson (2013) scored the JRAS alongside the JSORRAT II and two adult instruments on 636 juveniles who had sexually offended. Recidivism was tracked over two time periods: before adulthood (age eighteen) and afterwards. Findings showed that the adult tools were able to predict all types of juvenile recidivism at the same level of accuracy as the juvenile ones. The predictive validity of the JRAS and the other tools in predicting adult sexual recidivism was substantially lover than the predictive accuracy achieved in predicting juvenile sexual recidivism.
•A decision by the New Jersey Supreme Court was the impetus for the development of a risk assessment scale for juveniles. The JRAS was based on the Registrant Risk Assessment Scale (RRAS) for adults who have offended (Ferguson, Eidelson and Witt, 1998; Witt et al., 1996).
Juvenile Risk Assessment Scale (JRAS)
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
Age Appropriateness
•The items include static and dynamic variables and are sub divided into three broad areas: (1) sex offence history, (2) antisocial behaviour and (3) environmental characteristics.
Author / Publisher Hiscox, Witt, and Haran
•Manual available at: www.nj.gov/oag/dcj/megan/jras manual scale 606.pdf
•Validation studies on the JRAS have been based on low risk samples which may contribute to the lack of predictive accuracy in relation to recidivism (Hempel et al., 2013).
General Notes
RATED page updated: August 2019 © Risk Management Authority 2019
•No validation with females.
•There is some crossover with internet offending, for the possession of child pornography counts as one offence. The victims of internet offences, however, are not scored as victims on the JRAS.
•The JRAS is not designed to be used by younger children, adults or females (Rich, 2009).
RATED page updated: August 2019 © Risk Management Authority 2019
Name of Tool
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
Author / Publisher Rich Year 2017
•The J RAT is a 97 item structured clinical tool designed to aid clinical assessment of adolescent males who have or are alleged to have engaged in sexually abusive behaviour. Version 4 was published in 2017. Previous versions were published in 2001, 2003 and 2012.
•The tool has three scales used to measure and assess: (a) sexual risk, (b) risk for non sexual problematic behaviours and, if applicable, (c) risk for sexual behaviour that is non abusive but troubled or an area of concern.
Assessor Qualifications
Description
•The tool is designed to assess the likelihood or potential for sexual recidivism. The presence of risk items are listed as concerns, ranging from none to significant (Rich, 2009).
12 18
Juvenile Risk Assessment Tool Version 4 (J RAT)
•It also provides the clinician with a structured format for the assessment of risk, based upon factors frequently noted in current literature as relevant to risk of sexual recidivism.
Age Appropriateness
Tool Development
•The J RAT provides the evaluating clinician with a structured format for the assessment of risk, based upon factors frequently described in the professional literature and similar risk assessment instruments as relevant to the risk of sexual recidivism in juveniles.
Given that the tool includes items pertaining to mental health it is advisable that assessors should have training and experience in assessing risk within a youth mental health context. No further information pertaining to assessor qualifications.
•The J RAT has been in use since 2000, and is used across the United States. As a structured clinical instrument the J RAT is intended and designed to be part of a larger and more comprehensive psychosocial and risk evaluation of juveniles.
•The 97 items are grouped under 12 risk domains, each of which represents an overarching risk factor. Each risk domain represents an area of behaviour, capacity or skill, psychosocial functioning, cognition, relationships, or environmental conditions.
•A research project designed to measure the inter rater reliability of the J RAT has been underway since 2009. In addition, the study is designed to help pinpoint weaknesses in the design of the J
•The tool also contains 24 protective factors. These factors are also present in each of the 12 risk domains.
•For other information, please contact the author: philrich@philrich.ne
•Personal communication in early 2018 with the author revealed that the tool gets updated periodically but only in minor ways.
RATED page updated: August 2019 © Risk Management Authority 2019
The CI/J RAT (Cognitively Impaired Juvenile Risk Assessment Tool) is for adolescent males aged 12 18 with neurological or cognitive impairments such as autism. There is also the possibility of utilising this tool with adolescents with an IQ of fewer than 80 at the discretion of the assessor.
•The J RAT is not a statistically based assessment instrument. It is an organised method for the clinical assessment of risk for sexual re offense based on the professional literature.
The IM RAT (Interim Modified Risk Assessment Tool) containing ten individual risk elements within thirteen risk domains to allow for the on going assessment of an adolescent’s progress in treatment.
•Access to the J RAT assessments via the following websites: http://www.stetsonschool.org/risk assessment instruments.html http://www.philrich.net/risk assessment instruments.html
•The JRAT mandates written comments to explain the assigned risk in each domain, in addition to a written concluding narrative to justify the overall assigned risk level. Once a risk level has been assigned, the JRAS also defines characteristics from risk scenarios and factors relating to the behaviour of the adolescent (Rich, 2009).
General Notes
RAT, and help produce a stronger instrument. The J RAT has been re designed based on the first stage of the study and data analysis; it is now in use as Version 4 (Rich, personal communication, January 2013).
•There are a number of variations on the JRAS for specific groups (Rich, 2009):
The LA SATT, (Latency Age Sexual Adjustment Assessment Tool) designed to assess adolescent males aged between 8 and 13 years who have or are alleged to have engaged in sexually inappropriate behaviour.
Age Appropriateness
Name of Tool Juvenile Sexual Offender Recidivism Risk Assessment Tool II (JSORRAT II)
Author / Publisher Epperson
12 18 Assessor Qualifications
Year 2005 Description
•The JSORRAT II is 12 item actuarial tool designed to assess the risk of juvenile sexual recidivism among male juveniles who are aged between 12 to 18 years at the time of their index sexual offence.
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
Assessors must undertake a certified workshop where they will learn about the development of the JSORRAT II, how to score the instrument using a variety of case studies and also how to interpret its findings.
RATED page updated: August 2019 © Risk Management Authority 2019
•Ralston, Epperson and Edwards (2016) applied the JSORRAT II to 529 male adolescents who had sexually offended in Iowa. The predictive accuracy of the tool was found to be significant with an AUC of .70. In a breakdown of age groups, the J SORRAT II performed well for those aged 11 13 and 14 15, generating AUCs of above 0.70. The result for those aged 16 17 years, however, was not
•The JSORRAT II was developed through the identification of key predictors of sexual offending in a sample of 636 juveniles who had been charged for a sexual offence (Epperson et al., 2005).
•The tool is comprised solely of static items which include the youth’s sexual and non sexual offence history and previous experiences of victimisation. The first six items document sexual offending and the remaining ones are focused on non sexual offending (Ralston et al., 2017).
•Epperson et al. (2005) found that the tool had reasonable accuracy in predicting further juvenile sexual recidivism (AUC = .89) and sexual recidivism both as a juvenile and as an adult (AUC = .79).
•Viljoen et al. (2008) the JSORRAT II had excellent inter rater reliability (ICC =.89). It did not, however, significantly predict sexual, non sexual, and violent recidivism, both during treatment and post discharge.
•The authors define juvenile sexual recidivism as any new sexual offence committed prior to the age of 18.
Tool Development
•In an initial validation study, Epperson and Ralston (2015) reported: "Reliability of scoring the tool across five coders was quite high (intra class correlation coefficient [ICC] = .96).
•Rasmussen (2017) found that the J SORRAT II was not predictive yielding an AUC of .57 in a sample of 130 adolescents.
RATED page updated: August 2019 © Risk Management Authority 2019
•Ralston et al. (2017) tested data from ‘documented but uncharged’ sexual offences in Iowa and Utah when scoring the J SORRAT II. The inter rater reliability was sound with ICCs of .96 for Utah and .97 for Iowa. The predictive accuracy for the full dataset including charged crimes was moderate at .70 for Iowa, .65 for Utah and .67 for the combined sample. When only the uncharged crimes were assessed, however, the AUC was lower at .68 for Iowa, .65 for Utah and .60 for the combined scale. The main issue is that data from ‘documented but uncharged’ sexual offences cannot be used in the first six items of the J SORRAT II, as these require an official charge.
•The authors have not authorized the use of the JSORRAT II for forensic purposes in any country (apart from two states in the US) until additional validation evidence exists (Ralston, personal communication, January 2013).
•For more information on the JSORRAT II assessment, please e mail the primary author, Douglas Epperson, at: dleppers@calpoly.edu
•The inter rater reliability of the J SORRAT II was found to be acceptable when applied to fifty cases by eleven coders. The ICC results ranged from .67 to 1.00, with sexual and violent recidivism generating ICCs of .70 and .62 respectively (Ralston, Epperson and Edwards, 2016).
•Training on the JSORRAT II is available via the Global Institute of Forensic Research: https://www.gifrinc.com/course/jsorrat ii/
•The authors caution the interpretation of results from the study conducted by Viljoen and colleagues (2008) on several grounds, including (but not limited to) training, sampling and design issues.
significant. When the individuals whose only violent offence was sexual in nature were removed from the sample, the AUC generated was not found to be significant at .57.
•The tool is currently under going validation in the US. It is validated for use in Utah, where it was originally developed, Iowa, Georgia and California. In California, the J SORRAT II is the mandatory risk assessment tool for any male juveniles who have committed sexual offences (Ralston, Epperson and Edwards, 2016).
•Additional research related to the tool’s ability to predict general violence and adult recidivism, largely relies on inclusion of persons who have re offended sexually before age 18 in the recidivism criterion. When juvenile sexual recidivists are, therefore, removed from the sample, the JSORRAT II does not do well at predicting either violent recidivism (juvenile or adult) or adult sexual recidivism. The authors caution against the use of the tool in this respect given the lack of validation evidence (Ralston, Epperson and Edwards, 2016).
General Notes
•The tool is not designed for adolescents younger than 12 or those whose only sexual offences occurred when they were younger than 12 years old (Ralston, Epperson and Edwards, 2016).
•Epperson and Ralston’s (2015) study on 636 juveniles in Utah found that the predictive accuracy for sexual and sexually violent recidivism was significant at .89 for both types. When this was cross validated on a sample of 566 adolescents, the AUC was significant at .65 for sexual and sexually violent offences.
•The tool was developed using a review of the available literature and previous clinical experience with adolescents and young adults who have sexually offended.
•It consists of twenty dynamic items relating to sexual interests, behaviours and individual features like problem solving. Each of these are coded as either protective, neutral or risk. Since these items apply to behaviours and circumstances in the past two months, it could have the potential to be useful as a measure of change over time.
No specific assessor qualifications.
Assessor Qualifications
Tool Development
•Once totalled, ratings fall into one of five categories ranging from ‘predominantly protective,’ needing little or no intervention, through to ‘predominantly risk,’ requiring significant intervention.
Year 2017 Description
Name of Tool Protective and Risk Observations for Eliminating Sexual Offense Recidivism (PROFESOR)
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
•The tool should be scored using multiple sources of information: interviews with individuals themselves and parents/caregiver; review of security information; results from tests or other measures.
Author / Publisher Worling
General Notes
12 25
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
•The rationale behind the PROFESOR was to have a measure that “simultaneously considers both protective and risk characteristics” for the purposes of informing decisions about treatment rather than predicting future risk. This decision was additionally driven by findings of the current research relating to the validity of risk prediction tools such as the ERASOR and the J SOAP II, as well as emerging findings from the literature that some factors measured in other tools are not relevant to young people. The absence of protective factors and the relatively narrow age range of other well known risk assessment tools was also a motivating factor for the development of the PROFESOR (Worling, 2017).
•The PROFESOR is a structured checklist examining protective and risk factors in adolescents and young adults who have sexually offended. It is applicable to individuals aged 12 25 years old.
•The PROFESOR is not to be used to predict risk of future sexual offending; rather, its purpose is to facilitate planning interventions that may help to facilitate healthy sexual relationships and, thus, reduce sexual recidivism.
•The five categories which an individual can be placed into are intended to guide the type and intensity of intervention required.
•The tool may also be used for adolescents who have downloaded or distributed child abuse images; although further research is needed on this (Worling, personal communication, January 2018).
•A simple scoring sheet has been developed for the PROFESOR to assist with the final categorisation and is available here: http://www.profesor.ca/downloads.html
•Further information about the PROFESOR may be found at the website: http://www.profesor.ca/
•Since the PROFESOR covers an age range of fourteen years, it is pertinent to be sensitive to developmental nuances and expectations when using it. An emotionally intimate friendship, for instance, will likely look different at age 22 from what it would at age 13.
RATED page updated: August 2019 © Risk Management Authority 2019
•The measure is guided by the Structured Professional Judgement (SPJ) approach and the Risk, Need, Responsivity (RNR) Principle (Andrews and Bonta, 2010). The conceptual foundation of the SHARP is that sexually harmful behaviour is linked with and dependent upon the sexual development of the young person (Richardson, 2009).
•The SHARP was developed in the UK and is a derivative of the former Risk Assessment Matrix (RAM) which was devised by the author of the SHARP. It was subject to empirical evaluation in respect of sexual re offending (see Christodoulidies et al., 2005).
No further information pertaining to assessor qualifications.
Author / Publisher Richardson Year 2009
12 19 Assessor Qualifications
•To date, there have been no validation studies on this measure; although the measure was not intended to predict risk for sexual recidivism (Richardson, 2009).
Description
Category Youth Assessment: Sexual Violence Risk (Awaiting Validation)
•The SHARP is intended for use by a range of professionals involved in the assessment or case management of youth who display sexually harmful behaviours such as psychiatrists, psychologists, social workers and criminal justice professionals
•The tool is appropriate for use in community and secure settings (Richardson, 2009).
RATED page updated: August 2019 © Risk Management Authority 2019
Name of Tool Sexually Harmful Adolescent Risk Assessment Protocol (SHARP)
Age Appropriateness
•The SHARP has been updated and is now a 62 item structured assessment tool that evaluates sexually harmful behaviour of male adolescents aged between 12 and 19 years.
•The tool is appropriate for use with young persons diagnosed with learning disabilities or other psychiatric disorders (Richardson, 2009).
•The formulation of risk of the individual is characterised as ‘low,’ ‘moderate’ and ‘high.’
The tool is intended to be utilised by a range of professional groups who are involved in the assessment and case management of sexually harmful young people.
Tool Development
•Items present in the tool have been derived from clinical and empirical knowledge of risk assessment and child and adolescent general development.
General Notes
•The SHARP acts as a guide for the case management process; however, it does not generate probabilities of reconviction or predict sexual reoffending.
•For more information, the author can be contacted via email: graeme.richardson@ntw.nhs.uk
RATED page updated: August 2019 © Risk Management Authority 2019
•Author advises that the tool is in use at three NHS forensic adolescent mental health services (Richardson, personal communication, January 2012).
Year 2019 Description
Name of Tool Technology Assisted Harmful Sexual Behaviour: Practice Guidance (2nd edition)
RATED page updated: August 2019 © Risk Management Authority 2019
•There are three stages to the process: information gathering, case formulation and safety and intervention planning (Allotey and Swann, 2019).
Age Appropriateness
•The second stage of case formulation is broken down across nine areas: childhood (online and offline); adolescence (online and offline); neuropsychology; vulnerability; why now (e.g. looking at triggers, influences, etc ); facilitation (online and offline); harmful sexual behaviour; persistence (ongoing concerns, positive consequences the individual derives from TA HSB that could hinder them stopping); desistance (negative consequences the individual derives from TA HSB that could facilitate them changing their behaviour; strengths) (Allotey and Swann, 2019).
•The third and final stage involves safety and intervention planning utilising the results from the case formulation. Professional hypotheses should be advanced with regards to which factors pertaining to the individual, their family and networks will promote and hinder future safety. These should inform safety planning (including supervision and monitoring) and the appropriate interventions to encourage desistance (Allotey and Swann, 2019).
Adolescent males ages 12 18 years.
Assessor Qualifications
•The TA HSB Practice Guidance provides a framework to structure clinical judgment and formulation around technology assisted harmful sexual behaviours in adolescents (Allotey and Swann, 2019).
•It is not an assessment model intended to measure recidivism; rather, it is a consensus based tool guiding practitioners’ judgment (Swann 2018, personal communication).
Category Youth Sexual Violence (Awaiting Validation)
Author/Publisher Allotey and Swann/The AIM Project in partnership with NSPCC
•In the first stage, information should be gathered from across a range of sources: interviews with young person (a minimum of three is recommended); interviews with parents and/or carers; discussions/meetings with relevant professionals such as police, health, social care practitioners; access to relevant evidence, e.g. text/online chat transcripts, victim interviews where applicable; access to other relevant documentation, such as care plans and incident reports. Infomration should relate to four domains: TA HSB factors that may cause harm; developmental factors; family factors; environmental factors.
•As part of the training, worksheets are provided to facilitate all three stages, with suggested questions and guidance about how to approach each item (Allotey and Swann, 2019).
Domain 3: Family factors, which may be causal or influencing in continuing or cessation of sexual behaviours: parents/carers response; their ability to supervise and monitor, as well as their willingness to engage in interventions; the quality of relationship with primary attachment figure(s).
This can incorporate both offline and online aspects and may involve the use of technology alongside contact HSB (Allotey and Swann, 2019).
Domain 1: TA HSB factors that may cause harm to the self and/or others: developmentally inappropriate use of mainstream pornography; viewing, disturbing or producing indecent images of children; sexual harassment; grooming; relationship to victim(s) and characteristics of victim(s) where applicable; attitudes towards victims of harmful sexual behaviour; evidence of escalation of behaviours; association between technology assisted harmful sexual behaviour and contact or non contact behaviours taking place offline; criminal history, antisocial attitudes or behaviours.
•There are currently no studies to validate the use of this guidance.
This guidance is to be used by experienced practitioners who have undertaken additional training in conduct HSB risk assessments (for example, AIM3, JSOAP II, ERASOR) and have also undertaken the TA HSB training developed to accompany this guidance (Allotey and Swann, 2019).
•The previous model published by the AIM Project in 2009 called the iAIM was designed to assist practitioners working with young people whose behaviour online was a cause for concern. In 2015, the AIM Project and NSPCC collaborated to examine the use of risk tools which focused on the use of technology in harmful sexual behaviour. It was agreed that the iAIM needed to be updated in line with technological advances (Allotey and Swann, 2019; The AIM Project 2019, personal communication)
•The areas covered in the guidance span four domains:
RATED page updated: August 2019 © Risk Management Authority 2019
Domain 2: Developmental factors relating to the wider context of the young person’s functioning and wellbeing and any history of abuse and/or trauma: social development, emotional wellbeing, trauma, misuse of alcohol and/or substances and physical/mental health issues.
Domain 4: Environmental factors in terms of: online activity; relationship with online environment and how this facilitated their TA HSB; availability of support services; quality of relationships with peers (Allotey and Swann, 2019).
•The TA HSB Practice Guidance was developed from a wide range of research including a literature review (Belton and Hollis, 2016) and NSPCC research (Hollis and Belton, 2017). The literature review examined the role of new technologies for young people engaging in harmful sexual behaviour. This looked at the range of TA HSB and the crossover of behaviours; the characteristics of those who engage in TA HSB as well as those who engaged in both online and offline HSB; the impact of TA HSB. The NSPCC research was a qualitative study of young people who had been referred to treatment because of their technology assisted harmful sexual behaviour (Allotey and Swann, 2019; Belton and Hollis, 2016; Hollis and Belton, 2017; Swann 2018, personal communication).
Tool Development
•The definition of TA HSB used is derived from the NSPCC research study: “One or more children/young people engaging in sexual discussions or acts using the internet and/or any image creating/sharing or communication device which are considered inappropriate and/or harmful to self and/or other given their age or stage of development” (Hollis and Belton, 2017).
RATED page updated: August 2019 © Risk Management Authority 2019
•It is recommended that assessments should be completed by a co working pair, given the complexities involved (Allotey and Swann, 2019).
•When gathering evidence, three areas need to be carefully considered: unlike other types of harmful sexual behaviour, TA HSB may leave a forensic trail; the information sharing agreements between agencies and any barriers to this, e.g. ongoing police investigations may restrict access to relevant documents; this may be a traumatising experience for the professionals involved, so support should be provided (Allotey and Swann, 2019).
•The TA HSB guidance is to be used by experienced practitioners with training in similar risk assessments. This guidance should be used to supplement the AIM3 tool in cases where there is direct contact or non contact harmful sexual behaviour where there is a technology assisted element. Without the technology assisted element, the practitioner would only use the AIM3. In instances where there only appears to be technology assisted harmful sexual behaviour only this practice guidance would be used (Allotey and Swann, 2019; The AIM Project 2019, personal communication).
•The TA HSB is designed to be used with adolescent males aged 12 to 18 years old. Practitioners are not precluded from using this guidance on females or individuals with learning disabilities; however, they would have to be aware of the research relating to these groups (Allotey and Swann, 2019).
General Notes
ERASOR
Booth, A., Stepanova, E., Hackett, S., Sutton, A., Hynes, K., Sanderson, J. and Rogstad, K. (2016) Harmful sexual behaviour in children: evidence for identifying and assessing risk in children and young people who display harmful sexual behaviour. London, UK: NICE Evidence Review. Access Here
Chu, C. M., Ng, K., Fong, J. and Teoh, J. (2011) ‘Assessing Youth Who Sexually Offended: The Predictive Validity of the ERASOR, J SOAP II, and YLS/CMI in a Non Western Context.’ Sexual Abuse: A Journal of Research and Treatment 24(2), 153 174. Access Here
Barra, S., Bessler, C., Landolt, M. A. and Aebi, M. (2018) ‘Testing the Validity of Criminal Risk Assessment Tools in Sexually Abusive Youth.’ Psychological Assessment 30(1), 1430 1443. Access Campbell,HereF.,
Edwards, R., Whittaker, M. K., Beckett, R., Bishopp, D. and Bates, A. (2012) ‘Adolescents who have sexually harmed: An evaluation of a specialist treatment programme.’ Journal of Sexual Aggression 18(1), 91 111. Access Here
Hersant, J. L. (2006). Risk assessment of juvenile sex offender reoffense. Unpublished doctoral dissertation. Carbondale, Illinois: Southern Illinois University. [Not accessible]
McCoy, W. K. (2007). Predicting treatment outcome and recidivism among juvenile sex offenders: The utility of the JSOAPII and ERASOR in an outpatient treatment program Unpublished doctoral dissertation. Hunstville, Texas: Sam Houston State University. [Not Morton,accessible]K.E. (2003). Psychometric properties of four risk assessment measures with male adolescent sex offenders. Unpublished Master’s thesis. Ottawa, Ontario: Carleton University. [Not Nelson,accessible]R.(2011).
Predicting Recidivism Among Juvenile Sex Offenders: The Validity of the ERASOR. Unpublished Thesis. Paper 13, Roger Williams University, Rhode Island. Access Here.
Rojas Mejia, E. Y. (2013), Violence risk assessment with youth who have sexually offended: A psychometric examination of the Violence Risk Scale: Youth Sexual Offender Version (VRS:YSO). Unpublished doctoral dissertation. Saskatoon, Saskatchewan: University of Saskatchewan. Access Here
Rajlic, G. and Gretton, H. M. (2010) ‘An examination of two sexual recidivism risk measures in adolescent offenders: The Moderating Effect of Offender Type.’ Criminal Justice and Behavior 37(10), 1066 1085. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
VALIDATED TOOLS
Edwards, R., Beech, A., Bishopp, D., Erikson, M., Friendship, C. and Charlesworth, L. (2005) ‘Predicting dropout from a residential programme for adolescent sexual abusers using pre treatment variables and implications for recidivism.’ Journal of Sexual Aggression 11(2), 139 155. Access Here
Chávez, V. (2010). Is everyone rated equal? An examination of factors related to sexual risk in ethnically diverse male adolescents who have sexually offended. Unpublished doctoral dissertation. Lincoln, Nebraska: University of Nebraska. [Not accessible]
2019 © Risk Management Authority 2019
Worling, J. R. and Langton, C. M. (2015) ‘A prospective investigation of factors that predict desistance from recidivism for adolescents who have sexually offended.’ Sexual Abuse 27(1), 127 142. Access Here
Skowron, C. (2004) Differentiation and predictive factors in adolescent sexual offending. Unpublished Doctoral Dissertation. Ottawa, Canada: Carleton University. Access Here.
Worling, J. R., Bookalam, D. and Littlejohn, A. (2012) ‘Prospective validity of the Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR).’ Sexual Abuse: A Journal of Research and Treatment 24(3), 203 223. Access Here
Aebi, M., Plattner, B., Steinhausen, H. C. and Bessler, C. (2011) ‘Predicting Sexual and Nonsexual Recidivism in a Consecutive Sample of Juveniles Convicted of Sexual Offences.’ Sexual Abuse 23(4), 456 473. Access Here
Barra, S., Bessler, C., Landolt, M. A. and Aebi, M. (2018) ‘Testing the Validity of Criminal Risk Assessment Tools in Sexually Abusive Youth.’ Psychological Assessment 30(1), 1430 1443. Access Barroso,Here.R.,Pechorro, P., Ramião, E., Figueiredo, P., Manita, C., Gonçalves, R. A. and Nobre, P. (2019) ‘Are Juveniles Who Have Committed Sexual Offenses the Same Everywhere? Psychometric Properties of the Juvenile Sex Offender Assessment Protocol II in a Portuguese Youth Sample.’ Sexual Abuse. Access Here.
J. R. (2004) ‘The Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR): Preliminary Psychometric Data.’ Sexual Abuse: A Journal of Research 16(3), 235 254. Access Here
Prentky, R. and Righthand, S. (2003) Juvenile Sex Offender Assessment Protocol (JSOAP II) Manual. Bridgewater, MA: Justice Resource Institute. Access Here
Martinez, R., Flores, J. and Rosenfeld, B. (2007) ‘Validity of the Juvenile Sex Offender Assessment Protocol II (J Soap II) in a Sample of Urban Minority Youth.’ Criminal Justice and Behavior 34(10), 1284 1295. Access Here
J SOAP II
Viljoen, J. L., Elkovitch, N., Scalora, M. J. and Ullman, D. (2009) ‘Assessment of Re offence Risk in Adolescents who have Committed Sexual Offences: Predictive Validity of the ERASOR, PCL:YV, YLS/CMI and Static 99.’ Criminal Justice and Behavior 36(10), 981 1000. Access
WorlingHere.,
Worling, J. R. and Curwen, T. (2001). Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR; Version 2.0). Ontario, Canada: Ontario Ministry of Community and Social Services. Access Here.
Chu, C. M., Daffern, M., Thomas, S. and Lim, J. Y. (2012) ‘Violence risk and gang affiliation in youth offenders: a recidivism study.’ Psychology, Crime & Law 18(3), 299 315. Access Here.
Martinez, R., Rosenfeld, B., Cruise, K. and Martin, J. (2015) ‘Predictive validity of the JSOAP II: Does accuracy differ across settings?’ International Journal of Forensic Mental Health 14(1), 56 65. Access Here.
RATED page updated: August
S., Prentky, R., Knight, R., Carpenter, E., Hecker, J. E. and Nangle, D. (2005) ‘Factor Structure and Validation of the Juvenile Sex Offender Assessment Protocol (J SOAP).’ Sexual Abuse 17(1), 13 30. Access Here.
MEGA♪: Epperson, D. L., Ralston, C. A., Fowers, D., DeWitt, J. and Gore, K. S. (2006) ‘Actuarial risk assessment with juveniles who offend sexually: Development of the Juvenile Sexual Offense Recidivism Risk Assessment Tool II (JSORRAT II).’ In Prescott, D. (Ed.), Risk assessment of youth who have sexually abused: Theory, controversy, and emerging strategies Oklahoma City, OK: Woods ‘N’ Barnes, 118 169 Access Here
Rajlic, G. and Gretton, H. M. (2010) ‘An Examination of Two Sexual Recidivism Risk Measures in Adolescent Offenders: The Moderating Effect of Offender Type.’ Criminal Justice and Behavior 37(10), 1066 1085. Access Here.
Schwartz Mette, R. A., Righthand, S., Hecker, J., Dore, G. and Huff, R. (2019) ‘Long Term Predictive Validity of the Juvenile Sex Offender Assessment Protocol II: Research and Practice Implications.’ Sexual Abuse Access Here
Wijetunga, C., Nijdam Jones, A., Rosenfeld, B., Martinez, R. and Cruise, K. R. (2018b) ‘Using the Juvenile Sex Offender Assessment Protocol Revised to Assess Psychopathy: A Preliminary investigation.’ Criminal Justice and Behavior 45(4), 483 502. Access Here.
Viljoen, J. L., Scalora, M., Cuadra, L., Bader, S., Chávez, V., Ullman, D., and Lawrence, L. (2008) ‘Assessing Risk for Violence in Adolescents Who Have Sexually Offended: A Comparison of the J SOAP II, J SORRAT II, and SAVRY.’ Criminal Justice and Behavior 35(1), 5 23. Access Here.
RATED page updated: August 2019 Management Authority 2019
© Risk
Ralston, C. A. and Epperson, D. L. (2013) ‘Predictive validity of adult risk assessment tools with juveniles who offended sexually.’ Psychological Assessment 25(3), 905 916. Access Righthand,Here
Viljoen, J. L., Mordell, S. and Beneteau, J. L. (2012) ‘Prediction of adolescent sexual reoffending: A meta analysis of the J SOAP II, ERASOR, J SORRAT II, and Static 99.’ Law and Human Behavior 36(5), 423 438. Access Here
Prentky R. A., Li N. C., Righthand S., Schuler A., Cavanaugh D. and Lee A. F. (2010) ‘Assessing risk of sexually abusive behavior among youth in a child welfare sample.’ Behavioral Sciences and the Law 28(1), 24 45. Access Here
Vilijoen, J. L., Gray, A. L., Shaffer, C., Latzman, N. E., Scalora, M. J. and Ullman, D. (2017) ‘Changes in JSOAP II and SAVRY scores over the course of residential, cognitive behavioural treatment for adolescent sexual offending.’ Sexual Abuse: A Journal of Research and Treatment 29 (4), 342 374. Access Here
Wijetunga, C., Martinez, R., Rosenfeld, B. and Cruise, K. (2018a) ‘The Influence of Age and Sexual Drive on the Predictive Validity of the Juvenile Sex Offender Assessment Protocol Revised.’ International Journal of Offender Therapy and Comparative Criminology 62(1), 150 169. Access Here
Yates, H. M. (2005) A Review of Evidence Based Practice in the Assessment and Treatment of Sex Offenders. Pennsylvania: Office of Planning, Research, Statistics and Grants. Access Here.
RATED page updated: August 2019 Management Authority
© Risk
Miccio Fonseca, L.C. (2005) Fonseca Inventory of Sex Offenders’ Risk Factors (FISORF) Professional Manual. San Diego, CA: Author. [Not accessible]
Miccio Fonseca, L. C. (2016a) ‘MEGA cross validation findings on sexually abusive females: implications for risk assessment and clinical practice.’ Journal of Family Violence 31(7), 903 911. Access Here
MiccioHere
Miccio Fonseca, L. C. (2016b, May) MEGA♪: Second cross validation findings on sexually abusive youth. Presentation given at the Annual conference of the California Coalition on Sexual Offending (CCOSO), San Diego, CA. [Not accessible]
Miccio Fonseca, L. C. (2017b, September) Issues in assessment of sexually abusive youth: Intimacy deficits and erotically related protective factors and sexually abusive youth. 22st International Conference on Violence, Abuse, & Trauma and the National Summit on Interpersonal Violence & Abuse, San Diego, CA. [Not accessible]
Epperson, D. L. and Ralston, C. A. (2015)’ Development and validation of the Juvenile Sexual Offender Recidivism Risk Assessment Tool II.’ Sexual Abuse: A Journal of Research and Treatment, 27(6) 529 558. Access Here
Miccio Fonseca, L. C. (2017c) ‘The anomaly among sexually abusive youth: The juvenile sex trafficker.’ Journal of Aggression, Maltreatment & Trauma, 26, 5, 558 572. Access Here
Miccio Fonseca, L. C. (2013) ‘MEGA: A New Paradigm in Risk Assessment Tools for Sexually Abusive Youth.’ Journal of Family Violence 28(6), 623 634. Access Here.
Fagundes, M. (2013) ‘DSM IV TR diagnoses and risk levels of sexually abusive youth.’ Perspectives: California Coalition on Sexual Offending Quarterly Newsletter, Summer/Fall, 1, 5 8, 13. Access Here.
Miccio Fonseca, L. C. (2009) ‘MEGA ♪: a new paradigm in protocol assessing sexually abusive children and adolescents.’ Journal of Child and Adolescent Trauma 2(2), 124 141. Access
Miccio Fonseca, L. C. (1996) ‘Comparative Differences in the Psychological Histories of Sex Offenders, Victims, and Their Families.’ Journal of Offender Rehabilitation 23(3 4), 71 83. Access Here
Miccio Fonseca, L. C. (2017d, September) The anomalies among juvenile sex offenders: Sexually violent & predatory sexually violent. 22st International Conference on Violence, Abuse, & Trauma and the National Summit on Interpersonal Violence & Abuse, San Diego, CA. [Not accessible]
Fonseca, L. C. (2010) ‘MEGA♪: An Ecological Risk Assessment Tool of Risk and Protective Factors for Assessing Sexually Abusive Children and Adolescents.’ Journal of Aggression, Maltreatment & Trauma 19(7), 734 756. Access Here
2019
Miccio Fonseca, L. C. (2017a, Fall) Innovative scientific advancement in risk assessment of sexually abusive youth. Perspectives: California Coalition on Sexual Offending (CCOSO), Quarterly Newsletter, 7 10. Access Here
Miccio Fonseca, L. C. (2018a) ‘Family Lovemap and erotically related protective factors. Special Issue: Risk assessment of sexually abusive youth.’ Journal of Child Sexual Abuse 27(8), 901 917. Access Here
Leonard, M. and Hackett, S. (2019) The AIM3 Assessment Model: Assessment of Adolescents and Harmful Sexual Behaviour. Stockport: AIM Project. [Not accessible]
Allotey,AIM3
Miccio Fonseca, L. C. (2019) Family Lovemap, protective factors: Sex, intimacy, and sexually abusive youth. In Ireland, J. L., Ireland, C. A. and Birch, P. (eds.) Violent and sex offenders handbook (2nd edition). Milton Park, Abingdon, Oxon, UK: Routledge, 114 127. Access Here.
Miccio Fonseca, L. C. and Rasmussen, L. A. (2009) ‘New nomenclature for sexually abusive youth: Naming and assessing sexually violent and predatory offenders.’ Journal of Aggression, Maltreatment & Trauma 18(1), 106 128.
J. and Swann, R. (2019) Technology Assisted Harmful Sexual Behaviour: Practice Guidance (2nd edition). Stockport: AIM Project. [Not accessible]
Rasmussen, L. A. L. (2017) ‘Comparing predictive validity of JSORRAT II and MEGA♪\ with sexually abusive youth in long term residential custody.’ International Journal of Offender Rehabilitation and Comparative Criminology 62(10), 2937 2953. Access Here.
Beech, A. R. and Ward, T. (2004) ‘The integration of etiology and risk in sexual offenders: A theoretical framework.’ Aggression and Violent Behavior 10(1), 31 63. Access Here.
Miccio Fonseca, L. C. and Rasmussen, L. A. (2014) ‘MEGA♪: Empirical support for nomenclature on the anomalies: Sexually violent and predatory youth.’ International Journal of Offender Therapy and Comparative Criminology 59(11), 1222 1238. Access Here.
AWAITING VALIDATION
AR RSPB
Miccio Fonseca, L. C. (2018b, Spring) Sexually abusive youth who are transgender. Perspectives: California Coalition on Sexual Offending (CCOSO) Quarterly Newsletter, 1, 14 17. Access Here.
Miccio Fonseca, L. C. (2017e) ‘The juvenile female sex trafficker.’ Journal of Aggression and Violent Behavior 35, 26 32. Access Here.
Carson, C. (2019) AIM Assessment Model for Children Under 12 years old with Problematic Harmful Sexual Behaviours (3rd edition). Stockport: The AIM Project. [Not accessible]
RATED page updated: August 2019 © Risk Management Authority 2019
Miccio Fonseca, L. C. (2018c) ‘MEGA : Empirical Findings on the Preternatural: Sexually Violent and Predatory Sexually Violent Youth.’ Journal of Child & Adolescent Trauma, 1 11. Access Here
Miccio Fonseca, L. C. and Rasmussen, L. A. (2018) ‘Scientific evolution of clinical and risk assessment of sexually abusive youth: A comprehensive review of empirical tools. Special Issue on risk assessment of sexually abusive youth.’ Journal of Child Sexual Abuse 27(8), 871 900. Access Here
Viljoen, J. L., Elkovitch, N., Scalora, M. J. and Ullman, D. (2009) ‘Assessment of Re offence Risk in Adolescents who have Committed Sexual Offences: Predictive Validity of the ERASOR, PCL:YV, YLS/CMI and Static 99.’ Criminal Justice and Behavior 36(10), 981 1000. Access Here.
Skowron, C. (2004) Differentiation and predictive factors in adolescent sexual offending. Unpublished Doctoral Dissertation. Ottawa, Canada: Carleton University. Access Here
© Risk Management Authority 2019
Curwen, T. (2011a). A framework to assist in evaluating children’s risk to repeat concerning sexual behaviour. In M. Calder (Ed.), Contemporary Practice with Young People who Sexually Abuse: Evidence based Developments. Russell House Publishing, Lyme Regis: Dorset, 263 291. Access Here.
ERASORHere
RATED page updated: August 2019
Curwen, T. and Costin, D. (2007) ‘Toward assessing risk for repeated concerning sexual behavior by children with sexual behavior problems: What we know and what we can do with this knowledge.’ In Prescott, D. S. (ed.), Knowledge & practice: Challenges in the treatment and supervision of sexual abusers. Oklahoma City, OK: Wood ‘N’ Barnes, 310 344. Access Here.
Chu, C. M., Ng, K., Fong, J. and Teoh, J. (2011) ‘Assessing Youth Who Sexually Offended: The Predictive Validity of the ERASOR, J SOAP II, and YLS/CMI in a Non Western Context.’ Sexual Abuse: A Journal of Research and Treatment 24(2), 153 174. Access Here
Curwen, T. (2011b) Current evidence regarding the utility of a framework to assess risk to repeat concerning sexual behaviour (AR RSPB). Unpublished manuscript. [Not accessible]
Edwards, R., Whittaker, M. K., Beckett, R., Bishopp, D. and Bates, A. (2012) ‘Adolescents who have sexually harmed: An evaluation of a specialist treatment programme.’ Journal of Sexual Aggression 18(1), 91 111. Access Here.
Barra, S., Bessler, C., Landolt, M. A. and Aebi, M. (2018) ‘Testing the Validity of Criminal Risk Assessment Tools in Sexually Abusive Youth.’ Psychological Assessment 30(1), 1430 1443. Access Campbell,HereF., Booth, A., Stepanova, E., Hackett, S., Sutton, A., Hynes, K., Sanderson, J. and Rogstad, K. (2016) Harmful sexual behaviour in children: evidence for identifying and assessing risk in children and young people who display harmful sexual behaviour. London, UK: NICE Evidence Review. Access Here
Curwen, T., Jenkins, J. and Worling, J. D. (2009) ‘Differentiating children with and without a history of repeated sexual behaviors.’ Journal of Child Sexual Abuse 23(4), 462 480. Access
Nelson, R. (2011). Predicting Recidivism Among Juvenile Sex Offenders: The Validity of the ERASOR. Unpublished Thesis. Paper 13, Roger Williams University, Rhode Island. Access Here
Rajlic, G. and Gretton, H. M. (2010) ‘An examination of two sexual recidivism risk measures in adolescent offenders: The Moderating Effect of Offender Type.’ Criminal Justice and Behavior 37(10), 1066 1085. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
JRAS
Worling, J. R. and Curwen, T. (2001). Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR; Version 2.0). Ontario, Canada: Ontario Ministry of Community and Social Services. Access Here.
Hiscox, S. P., Witt, P. H. and Haran, S. J. (2007) ‘Juvenile Risk Assessment Scale (JRAS): A predictive validity study.’ Journal of Psychiatry & Law 35(4), 503 539. Access Here
Witt, P. H., DelRusso, J., Oppenheim, J. and Ferguson, G. (1996) ‘Sex offender risk assessment and the law.’ Journal of Psychiatry & Law 24(3), 343 377. Access Here
Worling, J. R. and Langton, C. M. (2015) ‘A prospective investigation of factors that predict desistance from recidivism for adolescents who have sexually offended.’ Sexual Abuse 27(1), 127 142. Access Here.
(2009) Juvenile Sexual Offenders: A Comprehensive Guide to Risk Evaluation. New Jersey: John Wiley and Sons, Ltd. Access Here
JSORRAT Epperson,IID. L., Ralston, C. A., Fowers, D. and DeWitt, J. (2005) Development of a sexual offence recidivism risk assessment tool II (JSORRAT II). Unpublished manuscript. University of Iowa. [Not accessible]
Ferguson, G. E., Eidelson, R. J. and Witt, P. H. (1998) ‘New Jersey’s Sex Offender Risk Assessment Scale: Preliminary Validity Data.’ The Journal of Psychiatry & Law 26(3), 327 349. Access Here.
Caldwell, M. F., Ziemke, M. H. and Vitacco, M. J. (2008) ‘An examination of the sex offender registration and notification act as applied to juveniles: evaluating the ability to predict sexual recidivism.’ Psychology, Public Policy and Law 14(2), 89 114. Access Here
Rich,Here.P.
(2009) Juvenile Sexual Offenders: A Comprehensive Guide to Risk Evaluation. New Jersey: John Wiley and Sons, Ltd. Access Here.
Worling, J. R. (2004) ‘The Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR): Preliminary Psychometric Data.’ Sexual Abuse: A Journal of Research 16(3), 235 254. Access Here
J Rich,RATP.
Worling, J. R., Bookalam, D. and Littlejohn, A. (2012) ‘Prospective validity of the Estimate of Risk of Adolescent Sexual Offense Recidivism (ERASOR).’ Sexual Abuse: A Journal of Research and Treatment 24(3), 203 223. Access Here
Hempel, I., Buck, N., Cima, M. and Marle, H. van. (2013) ‘Review of Risk Assessment Instruments for Juvenile Sex Offenders: What is Next?’ International Journal of Offender Therapy and Comparative Criminology 57(2), 208 228. Access Here
Ralston, C. A. and Epperson, D. L. (2013) ‘Predictive validity of adult risk assessment tools with juveniles who offended sexually.’ Psychological Assessment 25(3), 905 916. Access
Ralston, C. A., Sarker, A., Philipp, G. T. and Epperson, D. L. (2017) ‘The impact of using documented but uncharged data on J SORRAT II predictive accuracy.’ Sexual Abuse: A Journal of Research and Treatment 29(2), 186 202. Access Here
Richardson, G. (2009) ‘Sharp Practice: The Sexually Harmful Adolescent Risk Protocol (SHARP).’ In Calder, M. C. (ed.) Sexual Abuse Assessments: Using and Developing Frameworks for Practice. Dorset: Russell House Publishing. Access Here.
Belton, E. and Hollis, V. (2016) A review of the research on children and young people who display harmful sexual behaviour online: what is developmentally appropriate online sexual behaviour, do children and young people with online versus offline harmful sexual behaviours (HSB) differ, and is there an association between online and offline HSB. London: NSPCC. Access Here
Ralston, C. A., Epperson, D. L. and Edwards, S. R. (2016) ‘Cross validation of the JSORRAT II in Iowa.’ Sex Abuse 28(6), 534 554. Access Here.
Hollis, V. and Belton, E. (2017) Children and young people who engage in technology assisted harmful sexual behaviour; a study of their behaviours, backgrounds and characteristics London: NSPCC. Access Here.
© Risk
TA Allotey,HSB J. and Swann, R. (2019) Technology Assisted Harmful Sexual Behaviour: Practice Guidance (2nd edition). Stockport: AIM Project. [Not accessible]
Viljoen, J. L., Scalora, M., Cuadra, L., Bader, S., Chávez, V., Ullman, D., and Lawrence, L. (2008) ‘Assessing Risk for Violence in Adolescents Who Have Sexually Offended: A Comparison of the J SOAP II, J SORRAT II, and SAVRY.’ Criminal Justice and Behavior 35(1), 5 23. Access Here.
Andrews,SHARP D. A. and Bonta, J. (2010) The Psychology of Criminal Conduct (5th edition) Cincinnati, Ohio: Anderson Publishing. Access here.
Christodoulides, T. E., Richardson, G., Graham, F., Kennedy, P. J. and Kelly, T. P. (2005) ‘Risk assessment with adolescent sex offenders.’ Journal of Sexual Aggression 11(1), 37 48. Access Here
Worling,PROFESORJ.R. (2017) Protective + Risk Observations for Eliminating Sexual Offense Recidivism Access Here
RATED page updated: August 2019 Management Authority 2019
Rasmussen, L. A. L. (2017) ‘Comparing predictive validity of JSORRAT II and MEGA♪ with sexually abusive youth in long term residential custody.’ International Journal of Offender Rehabilitation and Comparative Criminology 62(10), 2937 2953. Access Here.
Epperson, D. L. and Ralston, C. (2015) ‘Development and Validation of the Juvenile Sexual Offence Recidivism Risk Assessment Tool II.’ Sexual Abuse 27(6), 529 558. Access Here.
b.Expertise in the area of violence against women in relationships
•An assessor is required to make a judgment relating to a psychiatric diagnosis on two occasions: the presence of a mental health disorder and a substance use problem (Kropp, Hart and Belfrage, 2005).
•Users should have relevant knowledge, training and experience in individual assessment with perpetrators and victims of spousal assault and intimate partner violence.
Category Intimate Partner Violence and Stalking (Validated)
•This was developed to help police make risk management decisions. The authors developed this tool in 2005 due to the fact that “the SARA may not be the optimal tool for use by police because it is relatively long and requires specific judgments about mental health, such as personality disorder” (Kropp and Hart, 2004).
Assessor Qualifications
•The instrument is appropriate for use with male and females aged 18 years and older.
Age Appropriateness
Name of Tool Brief Spousal Assault Form for the Evaluation of Risk Version 2 (B SAFER)
a.Training and experience in individual domestic violence assessments
•The B SAFER is a 10 item structured guide for the assessment and management of risk in adult males and females with a history of intimate partner violence (IPV).
18+
•The items contained within the instrument are divided into two main sections: participant’s history of intimate partner violence (Section 1) and the participant’s psychological and social functioning (Section 2). A third domain was added in the Second Version of the tool in 2010 called ‘Victim Vulnerability’ factors containing five risk factors relating to the victim (e.g. unsafe living situation and inadequate access to resources) (Storey and Strand, 2012; Svalin, 2018).
Year 2010 Description
Author / Publisher Kropp and Hart
•The instrument does not apply cut off scores to determine the nature or degree of risk posed by an individuals. It requires users to consider the risk to intimate partners if no intervention was taken, with risk rated as low, moderate or high (Kropp and Hart, 2004).
Assessors should meet the following minimal qualifications:
RATED page updated: August 2019 © Risk Management Authority 2019
Strengths
•Requires victim interview.
RATED page updated: August 2019 © Risk Management Authority 2019
•Kropp and Hart (2004) Inter rater reliability in the first version of the B SAFER was claimed to be ‘excellent. ’ As the study utilised qualitative analysis, however, there is no statistical score.
This tool was derived from the SARA and is grounded in professional and scientific literature on spousal violence (Kropp and Hart, 2004).
a)UK Research
None available at present.
Empirical Grounding
Inter Rater Reliability
•It is takes less time to complete making it less resource intensive. It also has removed some of the technical jargon from SARA relating to mental disorder (Kropp and Hart, 2004; Storey and Strand, 2012).
b)International Research
•Slavin et al. (2017) tested the second version of the B SAFER with three police employees conducting 23 pairwise B SAFER assessments. The lowest level of inter rater agreement was for intimate relationship problems and mental disorders. It was at its highest level for violations of court orders. Inter rater reliability was satisfactory for assessments of violent threats and thoughts and fair for item 3 ‘escalation.’ The decision was made not to speak to victims as part of this assessment, suggesting that the information gleaned from victim interviews is pivotal to the B SAFER risk assessment process. The authors believe that this missing information likely contributed to the lower inter rater agreement on certain items and the missing values rate.
•Thijssen and Ruiter (2011) using four of the ten items presented in the original B SAFER, the authors found poor to moderate inter rater reliability (ICC), ranging from .21 (‘Mental health Problems’) to .74 (‘General Criminality’). The authors attribute poor reliability to insufficient information contained within the client’s files. The items with good inter rated reliability were violent acts, general criminality and substance use problems.
•Au et al. (2008) men with histories of IPV scored significantly higher ratings on the original B SAFER than men with no history. B SAFER correctly classified 95% of the sample.
RATED page updated: August 2019 © Risk Management Authority 2019
None available at present.
General Predictive Accuracy
•Gerbrandij et al. (2018) tested the second version of the B SAFER on a sample of 158 low risk individuals. Weak, non significant predictive validity was found for both violent and stalking reoffending when considered in isolation. The B SAFER item ‘violations of course orders’ was a consistent predictor for stalking reoffending but not violence.
•Au et al. (2008) assessed the original B SAFER for IPV offending in Hong Kong. It was found that it had good ‘concurrent validity,’ with scores being able to measure the instances of IPV. The ‘construct validity’ of the tool in terms of its ability to classify IPV perpetrators from non perpetrators was also found to be strong.
•A doctoral thesis found that item 6 ‘General Criminality’ on the B SAFER was most strongly correlated with the severity of repeat violence and the likelihood of repeat IPV incidents in the current situation. Item 7 ‘Intimate Relationship Problems’ also strongly correlated with repeated intimate partner violence episodes (Svalin, 2018).
•A study by Loinaz (2014) found that the original B SAFER score predicted recidivism with an AUC of .76. ROC analyses found that the tool’s predictive validity was 70%. The authors cautioned that there are still some uncertainties with the tool, for there are a high number of false positives in the sample. They surmise that this may be related to the follow up period of 40 release cases for 15 months.
Validation History
•Belfrage and Strand (2012) found that practice in Stockholm of targeting resources at high risk cases affected predictive results for recidivism.
b)International Research
a)UK Research
•Storey et al. (2014) examined the use of the second version of the tool by Swedish police officers in assessing 249 IPV cases. It was found that police officers’ ratings of risk made by the tool predicted recidivism rates. The authors suggested that the B SAFER may be better suited to police officers.
Applicability: Mental Disorders
•Storey and Strand (2012) carried out a study involving police officers’ assessment of 52 women in Sweden and compared and contrasted the use of the original version of the B SAFER with a previous study (Belfrage and Strand, 2008) that focused on male perpetrators. The results showed that the total B SAFER scores were higher for men and that females were rated as being higher risk based on fewer risk factors. The authors hypothesise that the police officers may have considered additional risk factors to those on the B SAFER, given the lower number of risk factors present within female perpetrators. This suggests that the B SAFER may be insufficient for assessing the risk of IPV in female perpetrators.
b)International Research
Validation History
•Storey et al. (2014) maintained that B SAFER may be better suited to police usage than other tools such as SARA.
Validation History
•In some European countries, the B SAFER is known as the Police Version of the SARA (SARA PV) (Kropp and Hart, 2015).
RATED page updated: August 2019 © Risk Management Authority 2019
•The Canadian Association of Threat Assessment Professions recommends that in cases where the B SAFER is used (e.g. due to time constraints), the SARA tool should thereafter be utilised to gain a more in depth assessment of IPV cases. The inclusion of victim vulnerability factors in addition to those relating to the perpetrator in the second version of the B SAFER was noteworthy (Kropp and Hart, 2015).
None available at present.
Applicability: Females
a) UK Research
Contribution to Risk Practice
Authors state that the tool can be used with all adults who have committed IPV “regardless of gender or sexual orientation” (Kropp, Hart and Belfrage, 2005: 71).
Applicability: Ethnic Minorities
No evidence available at present.
Validation History
No evidence available at present.
Other Considerations
•A study by Nesset et al. (2017) looked at use of the tool during emergency visits by the police in potential IPV cases. Six out of the 15 items on the second version of the B SAFER tool were used as a basis for decision making about which actions to take: arresting the perpetrator and/or relocating the victim; no further action. It appeared that physical violence and substance abuse problems increased the odds of arrest; whilst mental health problems and the presence of children at the scene made it more likely that victims would be relocated. The authors maintained that the B SAFER enabled police to gather data and make decisions about how to manage the case in ‘real time.’ Results from research suggest that the inclusion of a victim interview should strengthen the accuracy of the tool.
•The B SAFER is more focused on martial relationships than other types of intimate partner relationships (Echeburúa et al., 2009). Intimate relationship problems in the B SAFER refer to failures to maintain an intimate partner relationship that is stable in nature (i.e. free from conflict and/or separations) (Petersson, Strand and Selenius, 2016).
•Assessors using the B SAFER are required to assess the risk of future intimate partner violence looking at different potential scenarios. This scenario is not as well developed as it is in SARAV3. It is surmised that this is related to a lack of reasoning about the situational aspects in relation to global risk. It is, therefore, recommended that greater weight is ascribed to situational factors in global risk assessments (Svalin, 2018).
•The tool has shown to be useful in differentiating sub types of those who have offended (non pathological and antisocial pathological), classifying 79% of the sample correctly (Loinaz, 2014).
•Future research should focus on the applicability of the tool to female perpetrators of IPV, including additional risk factors to determine whether this may be more suitable to managing women who have committed IPV (Storey and Strand, 2012).
•Slavin et al. (2017) suggested that the best way to use this tool would be to educate police officers on personality and mental health factors, since psychiatric disorders have been strongly linked to lethal IPV cases.
RATED page updated: August 2019 © Risk Management Authority 2019
•Comparing antisocial (n=341) with family only (n=316) perpetrators showed that the antisocial ones displayed significantly more risk factors on the B SAFER. In particular, violent thoughts or threats as well as escalations of violent and threatful behaviour were linked to an increased risk for acute and severe or deadly intimate partner violence (Petersson, Strand and Selenius, 2016).
•The SARA helps characterise the risk an individual poses to his spouse, children, another family member, or any other person involved in terms of likelihood, imminence and severity. The instrument does not use actuarial or statistical methods to support decision making about risk; it is a structural professional judgment method offering guidelines for collecting relevant information and making decisions (Messing and Thaller, 2015)
•The SARA is a 24 item structured guide for spousal risk evaluations in individuals who are suspected of, or who are being treated for, spousal abuse. Eight items describe the nature of Intimate Partner Violence (IPV) in terms of the diversity, chronicity and escalation of behaviours as well as supervision violations. Ten items are coded on the perpetrator’s issues with social, interpersonal and psychological adjustments. Six items describe vulnerabilities that could interfere with a victim’s ability, opportunity or motivation to engage in self protective behaviour (Kropp and Hart, 2015).
•Access to clinical records (if applicable) and criminal justice case files are required. It is also necessary for the perpetrator and victim to take part in interviews (Messing and Thaller, 2015).
Age Appropriateness
Year 2015
Assessor Qualifications
Assessors are intended to have advanced training (Messing and Thaller, 2013) and experience with the victims or perpetrators of IPV (Kropp and Hart, 2015). Assessors should also keep abreast with updates in research relating to IPV (Kropp and Hart, 2015).
Name of Tool Spousal Assault Risk Assessment Guide Version 3 (SARAV3)
•In Step 1, evaluators gather and document basic case information. In the second step, they identify the presence of 24 factors across the three domains (nature of IPV, perpetrator factors and victim vulnerabilities). In Step 3, evaluators assess the relevance of factors in relation to the perpetration or prevention of future IPV. In the fourth step, evaluators describe the most likely scenarios of future IPV. In Step 5, evaluators recommend ways to manage IPV risk considering the factors and scenarios identified. In the sixth and final step, evaluators document their judgment regarding the overall level of risk (Kropp and Hart, 2015).
Author / Publisher Kropp and Colleagues
Kropp18+ and Hart (2015) suggest that the SARAV3 may be of assistance for evaluating adolescents aged between 15 to 18 years; however, research relating to its effectiveness with this age group is limited.
RATED page updated: August 2019 © Risk Management Authority 2019
Description
Category Intimate Partner Violence and Stalking (Validated)
None available at present.
Inter-Rater Reliability
•A third version of the tool was developed to include updates to the empirical literature on IPV risk assessment. Version 3 of the SARA has addressed problems from the earlier version raised in feedback from users of the tool: the critical item ratings have been eliminated; three items referring to the supervision violations have been combined into one risk factor; clearer and more consistent definitions of past and recent are used. Vulnerability factors reflecting barriers to a victim’s ability, opportunity or motivation to engage in self protective behaviour have been included to help develop safety plans (Kropp and Hart, 2015).
b)International Research
•The SARA has been used internationally in Australia, the United Kingdom, Norway, Germany, the Netherlands, Hong Kong and Singapore (Messing and Thaller, 2013).
Empirical Grounding
RATED page updated: August 2019 © Risk Management Authority 2019
•The original version of the SARA is based on empirical and clinical literatures. Also covers static and dynamic factors drawn partly from the HCR 20. Developments in the literature since the original SARA was published have been included in the third version of the tool.
•Belfrage et al. (2012) found low to moderate reliability of the ratings between the first and second contact with police for three of the items within the SARA ranging from .56 to .68.
Strengths
•The Spousal Assault Risk Assessment is a set of guidelines based on the structured professional judgement approach to risk assessment and it is designed to ensure that appropriate risk assessment is conducted for domestic violence. It consists of 20 items that have been identified from the literature as being of relevance to the likelihood of future domestic violence offending. Version 3 of the tool reflects advancements in IPV risk assessment technology and risk factors, particularly the importance of victim vulnerability (Kropp and Hart, 2015).
•A discretionary clinical over ride is available for situations that are not captured by the risk factors found in the tool.
•A broad definition of IPV is used, extending to violence in any intimate (i.e. sexual, romantic) relationship regardless of legal status or the gender of people involved. Violence in this context refers to actual, attempted or threatened physical harm (Kropp and Hart, 2015).
•Can be used with males and females, regardless of sexual orientation or culture (Kropp and Hart, 2015), as well as mentally disordered individuals (Wong and Gordon, 1999).
All of the research at present relates to previous versions of the SARA. Research on SARAV3 will be added as it becomes available.
a)UK Research
•Grann and Wedin (2002) found an excellent ICC value for the original SARA of .85 in a sample of 18 cases. Inter rater reliability was lower for part 1 (‘assessing general
violence) scores and part 2 (‘assessing spousal support’) scores (.74 and .88 respectively).
RATED page updated: August 2019 © Risk Management Authority 2019
General Predictive Accuracy
All of the research at present relates to previous versions of the SARA. Research on SARAV3 will be added as it becomes available.
•Grann and Wedin (2002) findings suggest marginal predictive ability for recidivism in the original SARA with AUCs ranging between .49 to .52 at the 6 month follow up period.
•Belfrage et al. (2012) in an 18 month follow up, the SARAV2 had moderate predictive accuracy (AUC = .63). Higher numerical scores on the SARA were associated with recidivism.
•Nicholls et al (2013): "The [original] SARA research reports nine AUCs ranging from 0.52 0.65. The interrater reliability (IRR) for the SARA was excellent for total scores, good for the summary risk ratings, and poor for the critical items."
•Olver and Jung (2017) found that SARA scores showed incremental validity and the psychological adjustment domain of SARA contributed to the prediction of IPV.
•Helmus and Bourgon (2011) previous studies relating to the original SARA found low to high accuracy between composite scores and intimate partner violence (IPV) recidivism (AUC = .59 .77) and violent recidivism (AUC = .58 .64). Similar predictive accuracy was found for the summary risk ratings and IPV (AUC =.56 .87) and violent recidivism (AUC = .55 .66).
•Llor Esteban et al. (2016) used the factors from the original version of the SARA as used in a Spanish population to classify men sentenced for IPV into groups ranked by high, medium and low risk.
•Jung and Buro (2017) tested a modified version of the SARA (consisting of 14 items) on 246 male perpetrators charged for IPV offences. Moderate predictive accuracy
Validation History
a)UK Research
b)International Research
None available at present.
a)UK Research
Validation History
Validation History
•In a study that included 43 females who had offended, the instrument was found to work equally well across genders. A correlation level of 37 (significant at the .005 level) was found with females using the SARA. It was found to have a greater ability to predict risk than other risk assessment tools (e.g. Domestic Violence Screening Inventory) (Hennepin County Department of Community Corrections and Rehabilitation Office of Planning, Policy and Evaluation, 2011).
No evidence available at present.
Applicability: Mental Disorders
was shown for predicting IPV behaviours: post index IPV charges and IPV convictions generated AUCs of .68 and .74 respectively.
Validation History
•The SARA has the ability to create an awareness of risk factors pertinent to the individual’s risk of reoffending.
Applicability: Ethnic Minorities
•Messing and Thaller (2013) reported an AUC of .628 and a K score of 6 in the original SARA.
No empirical evidence available at present.
SARAV3 contains a question about the presence of major mental disorders.
•The SARA can aid assessors in identifying risks and responsivity factors specific to the individual (e.g. criminal lifestyle, presence of mental health problems).
•The dynamic factors included in the SARA can contribute to the determination of the level of monitoring / rehabilitative efforts required.
Contribution to Risk Practice
Applicability: Females
None available at present.
RATED page updated: August 2019
© Risk Management Authority 2019
b)International Research
RATED page updated: August 2019 © Risk Management Authority 2019
•Observed fluctuation in inter rater reliabilities could be due to the fact that the SARA is not an actuarial measure. Moreover, the previous studies may have used incomplete records to score the SARA.
•The SARA can aid assessors in developing risk formulations and risk management strategies.
Other Considerations
•SARA’s validity is dependent on well trained professionals who possess the assessment skills needed to classify and identify individuals who have committed intimate partner offences that have apparent and known DV risk factors.
•Consideration should be given as to whether to use other SPJ tools in conjunction with the SARA.
•To address the time constraints of SARA, Kropp, Hart and Belfrage (2005) developed a shortened version of the SARA, the Brief Spousal Assault Form for the Evaluation of Risk (B SAFER). Much of the new literature about IPV risk assessment has been incorporated into the B SAFER (Kropp and Hart, 2015). There is a separate RATED entry for this tool.
•The original SARA and the Brief Spousal Assault Form for the Evaluation of Risk (B SAFER) are consistently cited in the literature (Campbell et al., 2003; Vitacco et al., 2012; Williams and Houghton, 2004) as credible tools for assessing risk of violence and establishing a prevention plan. While no instrument or process can perfectly predict the risk for intimate partner violence, these instruments provide a systematic way to assess risk for violence and re offense (Wilson and Goss, 2013)
•The SARA does not assess relationship status.
•Since the SARA was said to be more ‘user friendly’ and ‘adaptable’ than other risk assessment tools, it was adapted for use in South Africa (Londt, 2014).
•The developers of the SARA caution that if no information is available about a particular factor then this should be left uncoded (i.e. omitted from the assessment). When items are omitted, evaluators should take note of this and qualify their judgments accordingly, considering how their opinions may have changed if those item(s) had been completed (Kropp and Hart, 2015).
If there is a history of physical assault outside of intimate relationships, the evaluator should consider using the Historical Clinical Risk 20 Version 3 (HCR 20V3). For IPV cases involving sexual assault where the perpetrator has a history of sexual violence outside of intimate relationships, using the Risk for Sexual Violence Protocol (RSVP) should be considered. In cases where there is long term targeting of a victim following the end of a relationship, the evaluator should consider using the Stalking Assessment and Management (SAM) (Kropp and Hart, 2015).
•The DVI generates a percentile score for each of the scales to inform subsequent treatment interventions (Karca, personal communication, January 2012).
Category Intimate Partner Violence and Stalking (Awaiting Validation)
•The items are spread across six scales: (1) truthfulness, (2) violence (lethality), (3) control, (4) alcohol, (5) drugs and (6) stress coping abilities. DVI areas of inquiry were established after extensive review of domestic violence literature (DVI.com).
Description
•Lindeman and Khandaker (2011) used a large scale sample of 18, 770 individuals who had committed IPV offences and found that all DVI scores substantially exceeded the reliability coefficient. Males scored higher than females on the following scales: control, alcohol, drugs, stress coping abilities and truthfulness. Those with multiple IPV offences scored significantly higher than those with first time offences on all subscales apart from the ‘truthfulness’ scale.
Name of Tool Domestic Violence Inventory (DVI)
Year 2006
Tool Development
RATED page updated: August 2019 © Risk Management Authority 2019
No specific user qualifications stated in relation to use of the tool; although users are not typically clinicians or diagnosticians.
•The DVI is a 155 item self report actuarial assessment intended for use with persons accused or convicted of domestic violence or related offences.
Age Appropriateness
•The DVI scales measure the criminogenic needs that contribute to IPV (Lindeman and Khandaker, 2011).
Author / Publisher Lindeman
Assessor Qualifications
•An unpublished thesis (Herndon, 2014) found that although DVI percentile scores did not significantly predict reoffending status, other variables like probation outcome, sentence served and educational attainment did.
•Risk of the client is categorized as either ‘low’ (0 39%), ‘medium’ (40 69%), ‘problem’ (70 89%) or ‘severe problem’ (90 100%). A problem is not identified until a scale score is at the 70th percentile or higher. Severe problems represent 11% of those evaluated with the DVI.
Adult males and females.
Derivatives of the assessment (i.e. DVI Juvenile) can assess risk from the age of 12 upwards.
•Paper pencil administration takes on average 30 minutes and tests are scored electronically. Tests are all computer scored since that is said to be “objective, accurate and fair” (DVI.com).
General Notes
DVI Short Form is a brief (76 items) version of the DVI designed for use with those who are reading impaired. It takes about 15 20 minutes to complete http://www.dvi short form.com/DVISF.html
•A study by Bouffard and Muftić (2007) found that lower scores on DVI sub scales were linked to a lower potential for IPV reoffending. Males recommended for treatment received higher scores on the alcohol, control, violence and stress coping scales.
•For further information, please visit the following website: http://www.domestic violence inventory.com/
•There are three derivatives of this instrument which include:
•Small to large correlations were found for the six subscales in relation to domestic violence arrests (Lindeman, 2006).
•Pike and Buttell (2003) carried out a 12 month follow up using a sample of 100 males attending a batterer intervention program Significant differences were observed between the DVI pre and post treatment scores on psychological variables related to domestic violence. No significant differences in scores were found between African American and Caucasian males.
DVI Juvenile is designed for use with individuals aged between 12 and 17 years in court, probation and treatment settings. It consists of 149 true false and multiple choice items and takes around 25 30 minutes to complete. The tool has been standardised on both male and female juveniles
•Unpublished studies have collated data on the validity of the subscales (see ‘Online Testing’ website).
More information available at: http://domestic violence inventory juvenile.com/index.html
DVI Pre Post is a tool consisting of 147 items and is designed to examine pre and post treatment changes in the attitudes and behaviour of the individual in relation to domestic violence by administering it before and after treatment and comparing scores. Further information is available from: www.dvi pre post.com.
RATED page updated: August 2019 © Risk Management Authority 2019
Author / Publisher Hilton and Colleagues
Category Intimate Partner Violence and Stalking (Awaiting Validation)
•Rettenberger and Eher (2013) found the DVRAG yielded good predictive validity for domestic violence (AUC. 71), general criminal (AUC .70) and general violent reoffending (AUC .70)
•The DVRAG is a 14 item actuarial tool which assesses the probability of IPV perpetrated by males against a female partner (Rice, Harris and Hilton, 2010).
Tool Development
Year 2008
General Notes
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
•The DVRAG was developed from the ODARA through the addition of clinically relevant information that is not routinely available to the police (Hilton et al., 2008).
•A higher score on the DVRAG indicates a greater risk level. (Hilton et al., 2008; Hilton, Harris and Rice, 2010).
Description
18+ Assessor Qualifications
Name of Tool Domestic Violence Risk Appraisal Guide (DVRAG)
•Hilton et al. (2008) the DVRAG achieved good predictive validity (ROC = .71). DVRAG scores correlated significantly with recidivism (r =.30), number of recidivistic offences (r =.37), severe physical abuse (r =.37) and total injury (r =.39). The DVRAG scores exhibited excellent inter rater reliability (r = .92).
•The tool was created as a complement to the Ontario Domestic Assault Risk Assessment (ODARA), combining the ODARA item scores with total score on the Psychopathy Checklist Revised (PCL:R) (Hilton et al., 2008)
•The DVRAG was developed from the ODARA (see p. 76) through the addition of clinically relevant information that is not routinely available to the police (Hilton et al., 2008).
The DVRAG is intended for use by forensic clinicians and criminal justice officials who can access in depth information.
•It is recommended that this tool is used only when the assessor has access to detailed clinical and/or correctional data.
RATED page updated: August 2019 © Risk Management Authority 2019
•The manual for the DVRAG system is contained in the book: Hilton, N.Z., Harris, G.T., and Rice, M.E. (2010). Risk assessment for domestically violent men: Tools for criminal justice, offender intervention, and victim services. Washington, DC: American Psychological Association (Access Here).
•Rettenberger and Eher (2013) the DVRAG displayed moderate accuracy in predicting domestic violence recidivism (AUC = .71) in a sample of 66 high risk males who had committed sexual offences against their current or former partners.
•Trinh (2010) found large effect sizes between scores on the DVRAG and domestic violence recidivism
•Hilton et al. (2008) the DVRAG achieved good predictive validity (ROC = .71). DVRAG scores correlated significantly with recidivism (r =.30), number of recidivistic offences (r =.37), severe physical abuse (r =.37) and total injury (r =.39). The DVRAG scores exhibited excellent inter rater reliability (r = .92).
Year 2006
RATED page updated: August 2019 © Risk Management Authority 2019
Description
Category Intimate Partner Violence and Stalking (Awaiting Validation)
Author / Publisher Williams and Grant
•The DVSI R is an 11 item actuarial risk assessment that examines the risk of intimate partner violence (IPV) in males. This is a screening measure that should be followed by a more comprehensive assessment of IPV risk (Nicholls et al., 2013).
•Seven of the items relate to behavioural history of the perpetrator and the remaining four items examine substance abuse, the use of objects as weapons, employment status and the presence of children during the offence.
•It is intended to assist clinicians in making pre arrangement recommendations based on the likelihood of reoffending (Messing and Thaller, 2015).
Name of Tool Domestic Violence Screening Inventory Revised (DVSI R)
The measure was initially developed for probation services.
Tool Development
•A review of recent studies relating to the DVSI and the DVSI R by Nicholls et al. (2013) indicated that the tool suggested that the tool has good internal consistency (α=.71); but the inter rater reliability was uncertain with this not being reported in any of the reviewed studies. Predictive
Age Appropriateness
No specified age range.
No further information pertaining to user qualifications.
It has been suggested that the tool is perhaps most appropriate for social workers within the court system to allow them to make decisions about pre trial release (Messing and Thaller, 2015).
•Risk factors included in the assessment were based on a thorough review of domestic violence literature and opinions made by consultations with judges, lawyers, victim advocates and police force personnel (Williams and Houghton, 2004).
•The DVSI R is a derivative of the original DVSI which was developed by the Colorado Department of Probation Services. The revised version was created due to the further modification of the DVSI items, as some were deemed to be confusing or redundant.
Assessor Qualifications
•The tool generates a total risk score and two summary scores which relate to the following: (1) imminent risk of violence to the victim of an incident; (2) imminent risk to another person known to the perpetrator (Williams, 2012). Risk is characterised as ‘low’, ‘moderate’ and ‘high’.
•Hennepin County Department of Community Corrections and Rehabilitation (2010) the SARA composite scores had better correlation with recidivism scores with female perpetrators than the DVSI (rs = .37 vs .20).
•The DVSI scores had higher correlation with recidivism scores of repeat offending compared to the SARA scores (rs = .20 vs .13). The revised version may thus encounter similar issues.
•The DVSI R has been found to measure risk across various types of relationships, aligning with the broadening of the definition of IPV in a number of states in the US (Williams, 2012).
•The tool is not designed to measure the risk of IPV becoming lethal (Campbell and Messing, 2017).
•The majority of validation research has been conducted by the authors of the tool.
•Previous research on the DVSI have suggested that there was a 39% chance of making a misclassification error (Wong and Hisashima, 2008).
•Williams and Grant (2006: 407) the findings of the initial validation study suggested that male perpetrators generally have higher scores than female perpetrators on the DVSI R.
RATED page updated: August 2019 © Risk Management Authority 2019
accuracy was also found to be moderate for serious threatening and physical violence, with AUCs ranging from .61 to .73
•Stansfield and Williams (2014) conducted a ROC analysis of 18 month follow up data from a sample of 29, 317 individuals in Connecticut, 70% of which were male. Survival analyses showed that the DVSI R predicted female recidivism over time, something which the authors claim shows promise for the use of DVSI R to predict IPV recidivism in female perpetrators. Ethnic minorities and males were more likely to be rearrested; the authors hence stress the need for instruments that are specific to gender and ethnicity.
•A study by Williams (2012) showed that the DVSI R had predictive accuracy across the five behavioural measures of recidivism and, bar one, this did not vary by gender, age or ethnicity. It was also found that the DVSI R composite score had moderate to high predictive accuracy (AUC) with all measures of recidivism including; new familial violent offences (.62) and protective restraining order violations (.72). Similar accuracy was noted across gender and ethnic minority groups.
General Notes
•Criminal justice case file information is required to use the measure (Messing and Thaller, 2013).
•Williams and Grant (2006) findings suggested that the DVSI R has moderate to high predictive accuracy (AUC) in relation to repeat IPV (.71) multiple versus single incidents of IPV (.79) imminent risk (.64) and risk to others (.61). DVSI R performed similarly in relation to the AUC values for different types of familial violence.
Tool Development
•Olver and Jung (2017) carried out survival analyses and found that the ODARA predicted IPV recidivism amongst both males and females. The tol also predicted general criminal and violent recidivism.
Author / Publisher Hilton and Colleagues
Age Appropriateness
•The ODARA was created in collaboration with the Ontario Provincial Police. Its purpose is for frontline use by police officers at the scene of an IPV incident or a follow up investigation (Rice, Harris and Hilton, 2010). The tool was developed to assist police decisions about detaining suspects and offering addition support to victims, and to assist court decisions about conditional release (Bowen, 2011).
Name of Tool Ontario Domestic Assault Risk Assessment (ODARA)
RATED page updated: August 2019 © Risk Management Authority 2019
•Higher scores on the ODARA also indicate that a suspect accused of assault will be more likely to commit more assaults, commit them within a short space of time, and cause more injury than a suspect with a lower score.
Initially designed for use within the police service.
Description
18+ Assessor Qualifications
•The tool was constructed from risk factors that were found to be statistically significant in predicting assault recidivism. These risk factors were identified via statistical analyses of follow up data from police and criminal justice datasets relating to men who had assaulted their female partner.
•The ODARA is a 13 item actuarial risk assessment instrument designed to assess the likelihood of domestic violence recidivism in males
•Criminal justice case file information is required to use the tool for criminal justice purposes; information only from a victim interview may be used to provide feedback to the victim
No professional assessor qualifications required; training to standard level of scoring reliability is encouraged.
Year 2004
•The ODARA is appropriate for use with males aged 18 years and older and has also been validated with females.
•It has been used in Canada, Austria, New Zealand, the UK and the US (Messing and Thaller, 2015).
Category Intimate Partner Violence and Stalking (Awaiting Validation)
General Notes
•Rice, Harris and Hilton (2010) found high ICC values of .95 following the selection of 10 cases and .94 during the follow up period. The ODARA also demonstrated moderate accuracy in predicting domestic violence recidivism in follow up periods as short as 6 months (ROC area =.64).
•Hilton et al. (2014) tested the ODARA on 30 female perpetrators of IPVagainst a current or former marital, common law or dating partner over a decade follow up period. They found that the measure predicted recidivism, with an ROC of .724.
•Jung and Buro (2017) tested the ODARA on 246 male perpetrators charged with IPV offences. It was found that this largely predicted reoffending, with AUCs of .75, .71 and .70 for general, violent and IPV recidivism respectively. An inter rater reliability of .91 was found when the ODARA was scored.
•A meta analysis by Messing and Thaller (2013) discovered that the ODARA had an average AUC of .666.
•Rettenberger and Eher (2013) report good predictive accuracy for domestic violence recidivism using the ODARA (AUC = .71). For general criminal and general violent recidivism, the AUC was between .66 and .71.
RATED page updated: August 2019 © Risk Management Authority 2019
•The ODARA was developed and tested only for ‘male to female’ assault(Rice, Harris and Hilton, 2010). A number of studies have looked at female to male assault and dating violence. Further details of studies relating to the ODARA can be found http://odara.waypointcentre.ca/Content/Resources/ODARA_101_Bibliography__pdf.pdfhere:
•Radatz and Hilton (2019) utilised the original ODARA construction and validation data of men with criminal charges for IPV offences (n=970) in order to test whether the ODARA can be used to guide treatment intensity decisions for those who have committed IPV offences. Results indicate that the low, medium and high categories of the ODARA can be used to inform a three tiered categorical system to advise on intervention programmes.
•Hilton and Eke (2016) used the ODARA on 93 men who had committed an IPV offence against a female partner (marital, cohabiting or dating). Recidivism was tested after seven and a half years, with an AUC of .67 being found for post index IPV reoffending. The ODARA also predicted post index stalking with an AUC of .78
•Hilton et al. (2008) found moderate predictive accuracy (ROC area =.67) in sample of individuals assessed as high risk of reoffending.
•Physical violence recidivism was found to be predicted by risk factors related to physical violent and antisociality (incarceration for more than 30 days). By contrast, risk factors for nonphysical abuse recidivism were more victim specific (e.g. victim concern, fear and threats to kill) and related to other crimnogenic needs (e.g. substance abuse). (Lauria et al., 2017)
•Seewald and colleagues (2017) compared the predictive accuracy of the ODARA compared to forensic psychiatrists using unstructured clinical judgment: AUCs were generated of 0.78 and 0.44 respectively.
•Hilton and Harris (2009) in a 5 year follow up, the ODARA attained moderate accuracy in distinguishing between recidivists and non recidivists (ROC Area = .74) when ambiguous cases of violent recidivism were removed from analyses.
•A study by Lauria et al. (2017) found that the ODARA predicted intimate partner violence physical assault and predicting the outcome of any further police contact for nonphysical intimate partner abuse, generating AUCs of .68 and .72 respectively. Moreover, total scores on the ODARA were able to differentiate between those who reoffended with a further physical assault and those who did not (AUC of .68), as well as individuals with and without any further non physical abuse (AUC=.72).
•ODARA training is available through an online, restricted access training programme: ODARA 101: The Electronic Training Program: http://odara.waypointcentre.ca/. This is estimated to take between 4 and 6 hours to complete. For further details, please e mail: odara@waypointcentre.ca
•A scientific and authorised translation of the ODARA into German was completed by Gerth et al. (2014)
•The manual for the ODARA system is contained in the book: Hilton, N.Z., Harris, G.T., and Rice, M.E. (2010). Risk assessment for domestically violent men: Tools for criminal justice, offender intervention, and victim services. Washington, DC: American Psychological Association (Access Here). A second edition is currently in progress.
RATED page updated: August 2019 © Risk Management Authority 2019
•The first structured risk assessment tool for stalkers, the SAM is designed to provide a systematic, standardised and practical framework for evaluating and managing risk in stalkers.
Description
Assessors are required to attend the relevant training course of around 20 hours. Case studies are included in the training.
•The 30 factors contained within SAM are divided into 3 domains: (1) Nature of Stalking, (2) Perpetrator Risk, and (3) Victim Vulnerability. The rationale behind this is to understand the motivations and seriousness of the stalking behaviour, as well as which type of stalker the perpetrator may be (McEwan et al., 2018; Storey, Hart and Lim, 2017). Assessors are encouraged to thereafter consider the various risk scenarios that may unfold in relation to nature, severity, imminence and frequency/duration (Belfrage and Strand, 2007).
•Storey et al. (2011) the SAM obtained average Kappa coefficients of .95 (range between .79 to .91).
RATED page updated: August 2019 © Risk Management Authority 2019
Assessor Qualifications
Year 2008
18+
Name of Tool Stalking Assessment and Management (SAM)
Author / Publisher Kropp, Hart and Lyon
•The instrument is not actuarial in nature and does not feature fixed scores or cut offs; overall ratings of low, medium and high are based on professional judgment.
•Belfrage and Strand (2007) found there to be a strong correlation between factors included in the SAM and the degree of risk: the more factors coded, the higher the risk for repeat stalking episodes.
The SAM is intended for use by criminal justice, security, and mental health professionals working in a variety of contexts where complaints of stalking arise.
Tool Development
Age Appropriateness
Category Intimate Partner Violence and Stalking (Awaiting Validation)
•This tool has undergone extensive pilot testing in Canadian and Swedish police sectors and mental health settings (Kropp, Hart and Lyon, 2008).
•Storey et al. (2009) the SAM obtained high intra class reliability values for the following; ‘nature of stalking’ subscale (.87), perpetrator risk factors (.81) and victim vulnerability scale (.77).
•The SAM builds on previous work on structured professional judgement approaches to violence risk assessment, which include the SARA guide and the RSVP. It offers a structured way to assess the vulnerability of the stalking victim (Storey, Hart and Lim, 2017).
General Notes
•Inter rater reliability of 39 cases was shown to initially be poor, due to confusion about what was meant by ‘stalking currency’ time frame, which could last from ‘several days’ to ‘many year.’ To resolve this issue, a consensus definition was adopted of a fixed timeframe of six months. This generated fair to moderate inter rater reliability for case prioritization (ICC=.77), future stalking (ICC=0.66) and serious physical harm (ICC=0.50) (Shea et al., 2018).
RATED page updated: August 2019 © Risk Management Authority 2019
•Foellmi et al. (2016) assessed 89 stalking perpetrators over a follow up period of two and a half years. The total and subscale scores were shown to predict recidivism; although the clinical risk ratings did not significantly do so. Inter rater reliability on the total score, nature and perpetrator subscales were found to be moderate, with ICC2 scores of .77, .64 and .88 respectively
•Risk assessment in stalking is still very much in its early stages (Binder, 2006).
•The SAM does not consider the perpetrator’s amenability or access to treatment options. The focus is on the stalking behaviour, as well as assessing the perpetrator’s motivations and the vulnerability of the victim (Foellmi et al., 2016).
•The SAM can aid assessors in developing risk formulations and risk management strategies. The ability to note ‘other considerations’ within the SAM allows for unusual risk factors to be taken into account to allow for a fuller consideration of the risk posed (Belfrage and Strand, 2007).
•Kropp et al. (2011) the instrument has good concurrent validity with other known risk assessments such as the VRAG and the PCL:SV. Their study also indicated fair to good inter rater reliability.
•Gerbrandij et al. (2018) tested the utility of the SAM on 158 low risk individuals. it was discovered that the SAM had weak, non significant predictive validity for stalking and violent recidivism. The items measuring distress and violations of supervision on the SAM were found to be a consistent predictor for stalking reoffences, but not violent reoffending, suggesting the tool may be better suited to predicting stalking rather than violence.
•The SAM was completed for 146 adult stalkers in a study by Shea et al. (2018). Case prioritization and risk for continued stalking items were shown to discriminate between high risk and low risk stalkers, with AUCs of .69 and .76. The total lifetime SAM scores also demonstrated moderate to good discrimination, with an AUC of .70.
•Storey et al. (2009) found an overlap between psychopathic traits on the PCL:SV and stalking behaviours captured on the SAM. It was suggested that the ‘coldness’ and ‘boldness’ present in stalkers had the potential to indicate psychopathy.
The SRP is intended primarily for use by mental health clinicians, however the tool can be used by other professionals working within law enforcement or other agencies who have completed the workshop and score the tool with diagnostic assistance from a clinician.
Name of Tool Stalking Risk Profile (SRP)
•If the victim of the stalking behaviour is a public figure, two other risk domains pertaining to escalation and disruption of the individual’s stalking behaviours are included.
Tool Development
RATED page updated: August 2019 © Risk Management Authority 2019
•Analogous to other validated SPJ tools like the HCR 20, the items within the assessment are scored according to the extent to which it is evidenced within the case.
The SRP, therefore, assesses multiple risk domains which include the risk of harm to others and the risk of psychosocial damage to the stalker (i.e. the risk that the stalker will experience psychological and/or social harm arising from their behaviours). The SRP assesses a range of risks relating to stalking: persistence, recurrence, stalking related violence and psychosocial damage to the perpetrator (McEwan et al., 2016).
Year 2009
•Upon completion of the assessment, the assessor will usually make three separate risk judgements about; (1) further stalking behaviour, (2) stalking related violence and (3) psychosocial damage to the stalker.
Description
Category Intimate Partner Violence and Stalking (Awaiting Validation)
•The tool is intended for use with males and females aged 18 and above.
Attendance at 2 day accredited training workshop is required to become an SRP assessor.
Author / Publisher Mackenzie, McEwan, Pathé, James, Ogloff and Mullen
•The tool is intended for use by clinicians involved in all aspects of stalking: assessing risks, defining management, intervention and treat strategies, preparing court reports and providing advice to victims (see the Stalking Risk Profile website).
Age Appropriateness
•Stalking risk is seen as a multi dimensional construct with a number of potential risk outcomes
•The SRP is a structured professional judgement tool used for assessing and managing risk in stalking cases
18+
Assessor Qualifications
•The SRP was developed in context to the lack of tools that effectively assessed the risk of individuals who engaged in stalking behaviours (McEwan, Pathé and Ogloff, 2011).
General Notes
•McEwan et al. (2018) maintain that the SRP is likely to be helpful in situations where assessors have access to a range of information about on going or past stalking episodes from a variety of sources.
•Use of the tool with individuals aged below 18 is discouraged due to evidence that motivations and risk factors may be different in juvenile stalking situations (McEwan, personal communication, June 2012).
•In a study of 241 stalkers, the inter rater reliability was found to be high for stalker type and moderate to substantial for risk judgments and domain scores (McEwan et al., 2018).
RATED page updated: August 2019 © Risk Management Authority 2019
•The motivation for the stalking behaviour is established at the beginning of the risk assessment. The context in which the stalking behaviour arose, the relationship between the victim and stalker and the role of mental illness are also assessed. It does not have a section focusing on victim vulnerability; although this is measured to some extent within the risk factors (Storey et al., 2009; Storey, Hart and Lim, 2017).
•The SRP uses the motivational typology proposed by Mullen, Pathé and Purcell (2009). Stalkers are divided into five typologies: rejected, resentful, intimacy seeking, incompetent and predatory, which are then used to group relevant risk factors for each domain
•For more information please visit the following website: www.stalkingriskprofile.com
•Validated risk assessments such as the HCR 20 and the LSI R would often over or under estimate the risk posed by the stalker (McEwan, Pathé and Ogloff, 2011).
Londt, M. P. (2014) ‘Batterer Risk Assessment: The Missing Link in Breaking the Cycle of Interpersonal Violence.’ The Social Work Practitioner Researcher 26(1), 93 116. Access Here.
VALIDATED TOOLS
RATED page updated: August 2019 © Risk Management Authority 2019
Messing, J. T. and J. Thaller. (2013) ‘The average predictive validity of intimate partner violence risk assessment instruments.’ Journal of Interpersonal Violence 28(7), 1537 1558. Access
Campbell, J. C., Webster, D., Koziol McLain, J., Block, C., Campbell, D., Curry, M. A. and Laughon, K.(2003) ‘Risk factors for femicide in abusive relationships: Results from a multisite case control study.’ American Journal of Public Health 93(7), 1089 1097. Access Here.
Kropp, P. R., Hart, S. D. and Belfrage, H. (2005) The Brief Spousal Assault Form for the Evaluation of Risk (B SAFER): User manual. Vancouver, British Columbia, Canada: ProActive ReSolutions Inc. Access Here
Llor Esteban, B., García Jiménez, J. J,. Ruiz Hernández, J. A. and Godoy Fernández, C. (2016) ‘Profile of partner aggressors as a function of risk of recidivism.’ International Journal of Clinical and Health Psychology 16 (1), 39 46. Access Here.
Grann, M. and Wedin, I. (2002) ‘Risk factors for recidivism among spousal assault and spousal homicide offenders.’ Psychology, Crime & Law 8(1), 5 23. Access Here.
Jung, S. and K. Buro. (2017) ‘Appraising Risk for Intimate Partner Violence in a Police Context.’ Criminal Justice & Behavior 44(2), 240 260. Access Here.
Helmus, L., and G. Bourgon. (2011). ‘Taking stock of 15 years of research on the Spousal Assault Risk Assessment Guide (SARA): A critical review.’ International Journal of Forensic Mental Health 10(1), 64 75. Access Here
Hennepin County Department of Community Corrections and Rehabilitation Office of Planning, Policy and Evaluation (DOCCR) (2011) Validation of Two Domestic Violence Risk Instruments: Domestic Violence Screening Instrument (DVSI) & Spousal Abuse Risk Assessment (SARA). Hennepin County, MN: Hennepin County Department of Community Corrections & Rehabilitation. [Not accessible]
Kropp, P., Hart, S., Webster, C. and Eaves, D. (1999) The Spousal Assault Risk Assessment (SARA) Guide: User’s Manual. Tornot, Canada: Multi Health Systems. [Not accessible]
Kropp, P. R. and Hart, S. D. (2015) The Spousal Assault Risk Assessment Version 3 (SARAv3) Guide: User’s Manual. Toronto, Canada: Multi Health Systems. [Not accessible]
Messing,Here J. T. and J. Thaller. (2015) ‘Intimate Partner Violence Risk Assessment: A Primer for Social Workers.’ British Journal of Social Work 45(6), 1804 1820. Access Here
H., Strand, S., Storey, J. E., Gibas, A. L., Kropp, P. R. and S. D. Hart. (2012). ‘Assessment and Management of Risk for Intimate Partner Violence by Police Officers Using the Spousal Assault Risk Assessment Guide.’ Law and Human Behavior, 36, 60 67. Access Here
Belfrage,SARA
Vitacco, M. J., Erickson, S. K., Kurus, S., and B. N. Apple. (2012). ‘The role of the Violence Risk Appraisal Guide and historical, clinical Risk 20 in U.S. courts: A case law survey.’ Psychology, Public Policy, and Law, 18(3), 361 391. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Echeburúa, E., Fernández Montalvo, J., de Corral, P. and López Goñi, J. J. (2009) ‘Assessing Risk Markers in Intimate Partner Femicide and Severe Violence: A New Assessment Instrument.’ Journal of Interpersonal Violence, 24(6), 925 939. Access Here
Belfrage, H. and Strand, S. (2012) ‘Measuring the outcome of structured spousal violence risk assessments using the B SAFER: Risk in relation to recidivism and intervention.’ Behavioural Sciences and Law 30(4), 420 430. Access Here.
Gerbrandij, J., Rosenfeld, B., Nijdam Jones, A. and Galietta, M. (2018) ‘Evaluating risk assessment instruments for intimate partner stalking and intimate partner violence.’ Journal of Threat Assessment and Management 5(2), 103 118. Access Here.
Olver, M. E. and Jung, S. (2017) ‘Incremental prediction of intimate partner violence: an examination of three risk measures.’ Law and Human Behavior 41(5), 440 453. Access Here
Wong, T. and J. Hisashima. (2008) State of Hawaii, 2003 2007 Domestic Violence Exploratory Study on the DVSI and SARA. Hawaii: Hawaii State Department of Health. Access Here
Nicholls, T. L., Pritchard, M. M., Reeves, K. A. and Hilterman, E. (2013) ‘Risk Assessment in Intimate Partner Violence: A Systematic Review of Contemporary Approaches.’ Partner Abuse 4(1), 76 168. Access Here
Belfrage, H. and Strand, S. (2008) ‘Structured spousal violence risk assessment: Combining risk factors and victim vulnerability factors.’ The International Journal of Forensic Mental Health 7(1), 39 46. Access Here
Kropp, P. R., Hart, S. D. and Belfrage, H. (2005) The Brief Spousal Assault Form for the Evaluation of Risk (B SAFER): User manual. Vancouver, British Columbia, Canada: ProActive Resolutions Inc. Access Here
Wilson, D. J. and Gass, W. M. (2013) “If I Can’t Have Her, No One Can”: Predicting the Risk of Intimate Partner Assault. Madison, WI: University of Wisconsin. Access Here.
Kropp, P. R. and Hart, S. D. (2004) The Development of the Brief Spousal Assault Form for the Evaluation of Risk (B SAFER): A Tool for Criminal Justice Professionals. Canada: Department of Justice, Research and Statistics Division. Access Here
Williams, K. R., and A. B. Houghton. (2004). ‘Assessing the risk of domestic violence re offending: A validation study.’ Law and Human Behavior 28(4), 437 455. Access Here.
Wong, S. and Gordon, A. (1999) The Violence Risk Scale 2nd Edition (VRS 2): Manual for the Violence Risk Scale. Saskatchewan, Canada: University of Saskatchewan. [Not accessible]
B Au,SAFERA.,Cheung, G., Kropp, R., Yuk chung, C., Lam, G. L. T. and Sung, P. (2008) ‘A Preliminary Validation of the Brief Spousal Assault Form for the Evaluation of Risk (B SAFER) in Hong Kong.’ Journal of Family Violence 23, 727 735. Access Here
Svalin, K. (2018) Risk Assessment of Intimate Partner Violence in a Police Setting: Reliability and Predictive Accuracy. Doctoral thesis. Malmo, Sweden: Malmö University. Access Here Storey, J. and Strand, S. (2012) ‘Assessing Violence Risk Among Female IPV Perpetrators: An Examination of the B SAFER.’ Journal of Aggression, Maltreatment and Trauma 22(9), 964 980. Access Here
Storey, J., Kropp, R. P., Hart, S., Belfrage, H. and Strand, S. (2014) ‘Assessment and management of risk for intimate partner violence by police officers using the BRIEF Spousal Assault Form for the Evaluation of Risk (B SAFER).’ Criminal Justice & Behavior 41(2), 256 271. Access Here
Loinaz, I. (2014) ‘Typologies, risk and recidivism in partner violent men with the B SAFER: a pilot study.’ Psychology, Crime & Law 20(2), 183 198. Access Here.
Petersson, J., Strand, S. and Selenius, H. (2016) ‘Risk Factors for Intimate Partner Violence: A Comparison of Antisocial and Family Only Perpetrators.’ Journal of Interpersonal Violence, 1 21. Access Here.
DVRAG
TOOLS AWAITING VALIDATION DVI
Bouffard, J. A. and Muftie, L. R. (2007) ‘An examination of the outcomes of various components of a coordinated community response to domestic violence by male offenders.’ Journal of Family Violence 22(6), 353 366. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Nesset, M. B., Bjørngaard, J. H., Nøttestad, J. A., Whittington, R., Lynum, C. and Palmstierna, T. (2017) ‘Factors Associated with police decisions on immediate responses to intimate partner violence.’ Journal of Interpersonal Violence, 1 18. Access Here
Lindeman, H. (2006). Domestic Violence Assessment: Validation of the Domestic Violence Inventory. [Not accessible]
Lindeman, H. and Khandaker, R. (2011) ‘Domestic violence inventory: introduction and standardisation in a large sample of domestic violence offenders.’ Family and Intimate Partner Violence Quarterly 4(1), 49 69. Access Here.
Slavin, K., Mellgren, C., Levander, M. T. and Levander, S. (2017) ‘The Inter Rater Reliability of Violent Risk Assessment Tools Used by Police Employees in Swedish Police Settings.’ Nordisk Politforskning 4(1), 9 28. Access Here
Herndon, K. E. Z. (2014) The Domestic Violence Inventory as a Tool for the Prediction of Domestic Violence [unpublished doctoral thesis]. Minneapolis, MN.: Walden University.
Buttell, F. P. and Pike, C. K. (2003) ‘Investigating the Differential Effectiveness of a Batterer Treatment Program on Outcomes for African American and Caucasian Batterers.’ Research on Social Work Practice 13(6), 675 692. Access Here
Thijssen, J. and de Ruiter, C. (2011) ‘Identifying Subtypes of Spousal Assaulters Using the B SAFER.’ Journal of Interpersonal Violence 26(7), 1307 1321. Access Here
Rice, M. E., Harris, G. T. and Hilton, N. Z. (2010) ‘The Violence Risk Appraisal Guide and Sex Offender Risk Appraisal Guide for Violence Risk Assessment and the Ontario Domestic Violence Risk Appraisal Guide for Wife Assault Risk Assessment.’ In Otto, R. K. and Douglas, K. S. (eds.) Handbook of Violence Risk Assessment. New York: Routledge, 99 119. Access Here.
Campbell,DVSI(R) J. and Messing, J. T. (2017) Assessing Dangerousness: Domestic Violence Offenders and Child Abusers (3rd edition). New York: Springer Publishing. Access Here
Stansfield, R. and Williams, K. R. (2014) ‘Predicting family violence recidivism using the DVSI R: Integrating survival analysis and perpetrator characteristics.’ Criminal Justice and Behaviour 41(2), 163 180. Access Here
Hilton, N. Z., Harris, G. T., Popham, S. and Lang, C. (2010) ‘Risk Assessment Among Incarcerated Male Domestic Violence Offenders.’ Criminal Justice and Behavior 37(8), 815 832. Access Here.
RATED page updated: August 2019
Hennepin County Department of Community Corrections and Rehabilitation Office of Planning, Policy and Evaluation (DOCCR) (2011) Validation of Two Domestic Violence Risk Instruments: Domestic Violence Screening Instrument (DVSI) & Spousal Abuse Risk Assessment (SARA). Hennepin County, MN: Hennepin County Department of Community Corrections & Rehabilitation. [Not accessible]
Messing, J. T. and J. Thaller. (2013) ‘The average predictive validity of intimate partner violence risk assessment instruments.’ Journal of Interpersonal Violence 28(7), 1537 1558. Access Here.
Trinh, B. V. (2010). A replication study of the Domestic Violence Risk Appraisal Guide (DVRAG) (Doctoral dissertation). Los Angeles: Alliant International University. Access Here.
© Risk Management Authority 2019
Williams, K. R. and Houghton, A. (2004) ‘Assessing the risk of domestic violence re offending: A validation study.’ Law and Human Behavior 28(4), 437 455. Access Here
Hilton, N. Z., Harris, G. T., Rice, M. E., Houghton, R. E. and Eke, A. W. (2008) ‘An In Depth Actuarial Assessment for Wife Assault Recidivism: The Domestic Violence Risk Appraisal Guide.’ Law and Human Behavior, 32(2), 150 163. Access Here
Messing, J. T. and J. Thaller. (2015) ‘Intimate Partner Violence Risk Assessment: A Primer for Social Workers.’ British Journal of Social Work 45(6), 1804 1820. Access Here
Rettenberger, M. and R. Eher. (2013) ‘Actuarial risk assessment in sexually motivated intimate partner violence.’ Law and Human Behaviour 37(2), 75 86. Access Here.
Messing, J. T. and J. Thaller. (2015) ‘Intimate Partner Violence Risk Assessment: A Primer for Social Workers.’ British Journal of Social Work 45(6), 1804 1820. Access Here.
Nicholls, T. L., Pritchard, M. M., Reeves, K. A. and Hilterman, E. (2013) ‘Risk Assessment in Intimate Partner Violence: A Systematic Review of Contemporary Approaches.’ Partner Abuse 4(1), 76 168. Access Here.
© Risk
Hilton,Here. N. Z. and A. W. Eke. (2016) ‘Non Specialization of Criminal Careers Among Intimate Partner Violence Offenders.’ Criminal Justice and Behavior 43(10), 1347 1363. Access Here.
ODARA
Jung, S. and K. Buro. (2017) ‘Appraising Risk for Intimate Partner Violence in a Police Context.’ Criminal Justice & Behavior 44(2), 240 260. Access Here.
Williams, K. R. (2012) ‘Family violence risk assessment: A predictive cross validation study of the Domestic Violence Screening Instrument Revised (DVSI R).’ Law and Human Behavior, 36(2), 120 129. Access Here.
Hilton, N. Z. and Harris, G. T. (2009) ‘How Nonrecidivism Affects Predictive Accuracy: Evidence From a Cross Validation of the Ontario Domestic Assault Risk Assessment (ODARA).’ Journal of Interpersonal Violence 24(2), 326 337. Access Here
RATED page updated: August 2019 Management Authority 2019
Hilton, N. Z., Harris, G. T., Rice, M. E., Houghton, R. E. and Eke, A. W. (2008) ‘An In Depth Actuarial Assessment for Wife Assault Recidivism: The Domestic Violence Risk Appraisal Guide.’ Law and Human Behavior, 32(2), 150 163. Access Here.
Wong, T. and J. Hisashima. (2008) State of Hawaii, 2003 2007 Domestic Violence Exploratory Study on the DVSI and SARA. Hawaii: Hawaii State Department of Health. Access Here.
Hilton, N. Z., Harris, G. T. and Rice, M. E. (2010) Risk assessment for domestically violent men: Tools for criminal justice, offender intervention and victim services. Washington, D. C.: American Psychological Association. Access Here
Hilton, N. Z., S. Popham, C. Long and G. T. Harris. (2014) ‘Preliminary validation of the ODARA for female intimate partner violence offenders.’ Partner Abuse 5(2), 189 203. Access Here
Williams, K. R. and S. R. Grant (2006) ‘Empirically Examining the Risk of Intimate Partner Violence: The Revised Domestic Violence Screening Instrument (DVSI R). ’ Public Health Reports 121(4), 400 408. Access Here
Messing, J. T. and J. Thaller. (2013) ‘The average predictive validity of intimate partner violence risk assessment instruments.’ Journal of Interpersonal Violence 28(7), 1537 1558. Access Here
Gerth, J., Rossegger, A., Urbaniok, F. and Endrass, J. (2014) ‘Das Ontario Domestic Assault Risk Assessment (ODARA) Validität und autorisierte deutsche Übersetzung eines Screening Instruments für Risikobeurteilungen bei Intimpartnergewalt [The Ontario Domestic Assault Risk Assessment (ODARA) Validity und Authorised German Translation of an Intimate Partner Violence Screening Tool].’ Fortschritte der Neurologie, Psychiatrie, 82(11), 616 626. Access
Lauria, I., McEwan, T. E., Luebbers, S., Simmons, M. and Ogloff, J. R. P. (2017) ‘Evaluating the Ontario Domestic Assault Risk Assessment in an Australian Frontline Police Setting.’ Criminal Justice and Behavior 44(12), 1545 1558. Access Here
Bowen, E. (2011) ‘An overview of partner violence risk assessment and the potential role of female victim risk appraisals.’ Aggression and Violent Behavior 16(3), 214 226. Access Here
Olver, M. E. and Jung, S. (2017) ‘Incremental prediction of intimate partner violence: an examination of three risk measures.’ Law and Human Behavior 41(5), 440 453. Access Here.
Radatz, D. L. and Hilton, N. Z. (2019) ‘Determining Batterer Intervention Program Treatment Intensifies: An Illustration Using the Ontario Domestic Assault Risk Assessment.’ Partner Abuse 10(3), 269 282. Access Here.
Binder,Here
Foellmi, M. C., Rosenfield, B. and Galietta, M. (2016) ‘Assessing risk for recidivism in individuals convicted of stalking offences predictive validity of the guidelines for stalking assessment and management.’ Criminal Justice and Behaviour 43 (5), 600 616. Access Here
Kropp, P. R., Hart, D. R. and Lyon, D. R. (2008) Guidelines for Stalking Assessment and Management: User Manual. Vancouver, Canada: ProActive Resolutions, Incorporated. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Messing, J. T. and J. Thaller. (2015) ‘Intimate Partner Violence Risk Assessment: A Primer for Social Workers.’ British Journal of Social Work 45(6), 1804 1820. Access Here.
R. L. (2006) ‘Commentary: The Importance of Professional Judgment in Evaluation of Stalking and Threatening Situations.’ Journal of the American Academy of Psychiatry and the Law 34 (4) 451 454. Access Here.
SAM
Rice, M. E., Harris, G. T. and Hilton, N. Z. (2010) ‘The Violence Risk Appraisal Guide and Sex Offender Risk Appraisal Guide for Violence Risk Assessment and the Ontario Domestic Violence Risk Appraisal Guide for Wife Assault Risk Assessment.’ In Otto, R. K. and Douglas, K. S. (eds.) Handbook of Violence Risk Assessment. New York: Routledge, 99 119. Access Here
Seewald, K., Rossegger, A., Urbanick, F. and Endrass, J. (2017) ‘Assessing the Risk of Intimate Partner Violence: Expert Evaluations Versus the Ontario Domestic Assault Risk Assessments.’ Journal of Forensic Psychology Research and Practice 17(4), 217 223. Access Here.
Belfrage, H. and Strand, S. (2009) ‘Validation of the Stalking Assessment and Management Checklist (SAM) in law enforcement: a prospective study of 153 cases of stalking in two Swedish police counties.’ International Journal of Police Science and Management 11(1), 67 76. Access
Rettenberger, M. and Eher, R. (2013) ‘Actuarial risk assessment in sexually motivated intimate partner violence.’ Law and Human Behaviour 37(2), 75 86. Access Here
McEwan, T. E., Shea, D. E., Daffern, M., MacKenzie, R. D., P., J. R. and Mullen, P. E. (2018) ‘The Reliability and Predictive Validity of the Stalking Risk Profile. Assessment 25(2), 259 276. Access Here
Shea, D. E., McEwan, T. E., Strand, S. J. M. and Ogloff, J. R. P. (2018) ‘The reliability and predictive validity of the Guidelines for Stalking, Assessment and Management.’ Psychological Assessment 30(1), 1409 1420. Access Here
Kropp, P. R., Hart, S. D., Lyon, D. R. and Storey, J. E. (2011) ‘The development and validation of the guidelines for stalking assessment and management.’ Behavioral Sciences & the Law 29(2), 302 316. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
McEwan, T. E., Shea, D. E., Daffern, M., MacKenzie, R. D., Ogloff, J. R. P. and Mullen, P. E. (2018) ‘The Reliability and Predictive Validity of the Stalking Risk Profile.’ Assessment 25(2), 259 276. Access Mullen,HereP.E., Pathé, M. and Purcell, R. (2009) Stalkers and their Victims (2nd Edition). Cambridge: Cambridge University Press. Access Here
McEwan, T., Pathé M. and Ogloff, J. R. P. (2011) ‘Advances in Stalking Risk Assessment.’ Behavioural Sciences and the Law 29(2), 180 201. Access Here.
Storey, J. E., Hart, S. D. and Lim, Y. L. (2017) ‘Serial stalking of mental health professionals: case preventability, analysis and formulation using the guidelines for stalking, assessment and management (SAM).’ Journal of Threat Assessment and Management 4(3), 122 143. Access Here
Storey, J. E., Gibas, A. L. Reeves, K. A. and Hart, S. D. (2011) ‘Evaluation of violence risk (threat) assessment training for police and other criminal justice professionals.’ Criminal Justice and Behavior 38(6), 554 562. Access Here
Storey J. E., Hart S. D., Meloy J. R. and Reavis J. A. (2009) Psychopathy and stalking. Law and Human Behavior 33(3):237 246. Access Here.
Storey J. E., Hart S. D., Meloy J. R. and Reavis J. A. (2009) Psychopathy and stalking. Law and Human Behavior 33(3):237 246. Access Here
Storey, J. E., Hart, S. D. and Lim, Y. L. (2017) ‘Serial stalking of mental health professionals: case preventability, analysis and formulation using the guidelines for stalking, assessment and management (SAM).’ Journal of Threat Assessment and Management 4(3), 122 143. Access SRPHere
Year 2003
Assessor Qualifications
•Multi agency use across Police and Probation, giving the benefits of a common language and understanding to the management of cases particularly in the context of MAPPA (Nicholls and Webster, 2014).
Description
•The instrument is comprised of three subscales: RM S, RM V and RM C. These provide an estimate of the long term likelihood of reconviction for sexual (RM S) or non sexual violent (RM V) offences.
Assessors must complete the appropriate training awarded by NOMS in order to use this tool.
Category Sexual Offending (Validated)
•Easy to score and interpret, yet training should be required for scoring accuracy and quality assurance measures introduced (Grubin, 2011).
Author / Publisher Thornton and Colleagues
•Each level parallels a statistical likelihood of reconviction: low (score of 1), medium (score of 2 or 3), high (score of 4 or 5) and very high (score of 6+).
Tool is designed for use with those involved in the risk assessment and management: police officers, social workers, probation officers and other practitioners.
Age Appropriateness
•The RM2000 is financially effective for services to use on a large scale, given it is light on resources and time (Pryboda, Tully and Browne, 2015).
Name of Tool Risk Matrix 2000 (RM2000)
18+
•The RM2000 is a statistically derived risk assessment tool for use with adult (18+) males convicted of sexual offences. At least one of these sexual offences should have been committed after the age of 16 (Wakeling, Howard and Barnett, 2011).
•The RM2000 appears to perform in a stable manner across the UK and generalisations can be made in Scotland (Grubin, 2011).
RATED page updated: August 2019 © Risk Management Authority 2019
The RM C is a combination of the scores obtained for RM S and RM V subscales. Two steps are involved in scoring the subscales. The first step looks at risk items: the number of occasions sentenced for a sexual offence, the number of occasions sentenced and age on release. The second step looks at four aggravating factors. The presence of 2 aggravating factors increases the risk category by 1 level; four of these raises the category of risk by 2 levels (Smid et al., 2014).
Strengths
a)UK Research
•Bengtson (2008) good inter rater reliability was found for the sexual offending subscale of the RM2000 (ICC = .72; k =. 85)
•The sexual and non sexual offence subscales generated moderate to high AUCs of .73 and .76 respectively in a sample of Scottish individuals who had committed sexual offences (Grubin, 2011).
Empirical Grounding
Inter Rater Reliability
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
•Barnett, Wakeling and Howard (2010) found moderate predictive accuracy for the RM S subscale (.68) and higher predictive accuracy for the RM V (.80) and RM C (.73) subscales.
b)International Research
•Knight and Thornton (2007) the RM2000 achieved an ICC of .82.
•It could be useful as screening mechanism in further assessments to allow for the allocation of resources.
a)UK Research
This tool was developed as a revision and improvement of the SACJ minimum (Structured Anchored Clinical Judgment Scale Minimum). Similar to the SACJ Min, the RM2000 utilizes a stepwise approach and is composed of two main scales (Grubin, 1998; Hanson and Thornton, 2000).
•Wakeling, Mann and Milner (2011) found excellent inter rater reliabilities (ICCs) in two studies ranging from .71 for Study 1 and .93 for Study 2. .
General Predictive Accuracy
Recent developments have explored the combining of RM2000 with Stable 2007 (Hanson et al., 2007), where the static and dynamic risk scales are joined together in a structured manner. Findings indicated that the STABLE 2007 generally added incremental predictive validity to the RM2000. (Helmus et al., 2015).
•Looman and Abracen (2009) found an ICC value for the RM2000 composite score of .81.
•Looman and Abracen (2010) the RM S and RM V subscales generated moderate to high predictive accuracy for sexual re offending in individuals convicted of rape offences (AUCs = .70 and .65 respectively).
•Craig et al. (2008) reported high levels of predictive accuracy in relation to non sexual violent reconvictions with AUCs with the .80 to .87 in a 10 year follow up study.
•Osborn et al. (2010) administered the RM2000 to 73 individuals convicted of internet offences; it was found that the tool overestimated the risk levels posed.
•Lehmann et al. (2016) found the RM2000 demonstrated moderate accuracy in predicting sexual, non sexual violent and violent recidivism in an international sample of 3144 individuals from the UK, Germany and Canada.
•Wakeling, Howard and Barnett (2011) found that the RM2000 scales had moderate to very good predictive accuracy ranging between .67 and .87. It should be noted, however, that the RM2000 overestimated the risk posed by those who had committed internet offences.
b)International Research
RATED page updated: August 2019 © Risk Management Authority 2019
•Webb et al. (2007) RM2000 significantly predicted formal failure (e.g. reconviction, breach/recall) for those who had engaged in child molestation offences (AUC = .71) and other sexually risky behaviours (AUC = .69); although in the case of internet offences, the RM2000 was only able to moderately predict drop out from treatment (AUC =.69).
•Parent, Guay and Knight (2011) examined the predictive accuracy of sexual recidivism for 590 individuals. The AUCs were .68, .52 and .62 for the RM2000/s, RM2000/v and RM2000/c respectively.
•Knight and Thornton (2007) moderate predictive accuracy with AUCs ranging between .63 and .67 in the follow up periods.
Applicability: Females
Validation History
•Kingston et al. (2008) the RM2000 attained moderate predictive accuracy with AUC values of .64 and .65 for the RM V and RM S subscales respectively.
No empirical evidence at present.
Contribution to Risk Practice
RATED page updated: August 2019 © Risk Management Authority 2019
a)UK Research
b)International Research
•The RM2000 is able to provide a brief scan of some static risk factors pertinent to the risk of sexual and violent recidivism and can highlight the need for further assessment of the individual’s risk of recidivism. RM2000 should be one component of a comprehensive and appropriate assessment package (Grubin, 2011)
•Lindsay et al. (2008) the RM2000 was unable to significantly predict recidivism in a sample of male adults with learning disabilities. Further research using the RM2000 was recommended, however, since it just fell short of significance.
•In their review of literature, Pryboda, Tully and Browne (2015) concluded that the use of the RM2000 was not supported as a measure of static risk for intellectual disabled individuals. It is suggested this may be attributable to the RM2000 failing to consider protective factors or any others related to desistance.
Validation History
Applicability: Ethnic Minorities
•The RM2000 is a useful tool to assign resources by predicting reconviction. It is used by the prisons, probation and the police in England and Wales (Smid et al., 2014).
No empirical evidence at present.
•Langton et al. (2009) In a sample of personality disordered individuals, only the RM V subscale predicted damage to property (AUC=.74).
•Tully and Browne (2015) argue that adding dynamic risk items would fit better with a rehabilitative approach to risk management and assessment for sexual offending. This would also provide a means by which to effectively plan treatment and evaluate individuals’ progress in treatment; however, difficulties remain in identifying and assessing dynamic risk factors of sexual offending.
Applicability: Mental Disorders
Validation History
Other Considerations
•The tool may be of limited use with first time offenders whose current offence may be unusual or contain sadistic elements (Beech et al., 2003; Craig, Browne and Stringer, 2004).
•An unpublished revision to the RM2000 scoring manual has made several changes that include, but are not limited to, the assessment of risk in older groups (aged 60 and over) and individuals who have committed non contact sexual offences (Thornton, 2010).
•The RM2000 is static in nature meaning in theory the final risk categories should remain the same over time and items cannot be targeted for change. The completion of the tool should be repeated when an individual moves between age categories, is convicted or cautioned due to a further offence or maintains a two year cohabiting relationship for the first time (Smid et al., 2014).
•For access to the manual, please visit the following website: www.birmingham.ac.uk/Documents/college les/psych/RM2000scoringinstructions.pdf
•An unpublished report by Howard and Wakeling (2019) examined whether the length of time without offending in the community affected contact sexual reoffending rates. Findings gave tentative support to reducing the risk by one category for every five years that an individual has been offence free in the community: for instance, if an individual was assessed as medium risk and upon release from prison into the community remained offence free for the next five years, it would be reasonable to reduce him to low risk.
•Studies have found that it overestimates the risks posed by those who had committed internet offences. A more accurate predictor of risk was found to be a revised version of the tool, RM2000R, which omits two aggravating factors: stranger victim and non contact offences (Osborn et al., 2010; Wakeling, Howard and Barnett, 2011).
•The instrument itself is normed on adult males with a previous sex offence history and is inappropriate for use with females, juveniles, and mentally disordered individuals.
•Helmus et al. (2015) found that the addition of Stable 2007 added incremental predictive validity to the RM2000. Internet offences are not counted as non contact offences if they are the only sexual offence.
•In 2014, National Offender Management Services advised of a revision to scoring. Data indicated that risk halves for five years in the community free of sexual offences (Thornton and Helmus, 2015). It is, therefore, recommended that those convicted of sexual offences who remain offence free in the community for five years or more should have their risk level reduced by one category. The term ‘offence free’ refers to no known criminal activity taking place, no convictions, no warnings or reprimands or breaches. It was also recommended that those age 60 and over should be put one level of risk on the RM2000 due to a decreased level of risk (Thornton and Helmus, 2015).
•It is recommended by Bryboda, Tully and Browne (2015) that the RM2000 should be used on conjunction with other validated assessment methods/ The authors also highlighted the importance of considering protective factors in relation to intellectual disabled individuals who commit sexual offences, something which the RM2000 does not currently do.
RATED page updated: August 2019 © Risk Management Authority 2019
•The RRASOR is a 4 item screening instrument for risk of sexual offence recidivism among males who have been convicted of at least one sexual offence.
Assessor Qualifications
•It is easy to score and interpret as it is quick and uses available and basic file information (Yates, 2005).
The author of this tool selected variables that have been found in previous meta analyses to have a minimum correlation of .10 with sexual recidivism in order to generate the four items (Hanson, 1997). These particular items are also loaded highest in Knight and Thornton’s (2007) Sexual Persistence and Male Victim Choice Factors.
•Empirically based on the 4 most robust risk factors found in the research about sexual offending. The developmental study found the RRASOR to have an ROC ranging between .62 and .77 (Hanson, 1997).
RATED page updated: August 2019 © Risk Management Authority 2019
Name of Tool Rapid Risk Assessment for Sex Offence Recidivism (RRASOR)
18+
Empirical Grounding
Inter Rater Reliability
•Easily coded with or without psychological tests/clinical assessment and does not require the individual’s participation.
Age Appropriateness
Category Sexual Offending (Validated)
Description
•The instrument relies on information obtained in files and has been tested extensively on Canadian and British forensic populations.
Expertise in assessing sexual violence risk and training on instrument.
Can be administered by a trained non clinician (Yates, 2005).
Author / Publisher Hanson Year 1997
Strengths
•Rettenberger et al. (2010) the RRASOR attained fair to acceptable predictive values for subgroups of sexual offences. For the rapist subgroup, the RRASOR attained moderate AUCs of .70 for general violent recidivism and .69 for general criminal recidivism. For the incest group, fair AUCs of .67 and .65 were generated for general violent and general criminal recidivism respectively. For the extra familial molest offending group, fair AUCs were generated of .64, .61 and .60 for sexual, general violent and general criminal recidivism respectively. Predictive accuracy was not predicted for sexual violent recidivism in this offending group, generated an AUC of .53.
•Langton et al. (2007) found excellent inter rater reliability for the RRASOR (r = .94).
•Hanson and Morton Bourgon (2009) a meta analysis study found the RRASOR to be an adequate predictor of sexual recidivism (average Cohen d = .60).
•Hanson (1997) the measure had been validated with a UK prison population in which the RRASOR attained moderate to strong AUC values (.61 .71).
General Predictive Accuracy
•Rettenberger et al. (2010) excellent inter rater reliability was found for the RRASOR with an ICC of .90
a)UK Research
•Knight and Thornton (2007) RRASOR generated an ICC of .82
Validation History
b)International Research
Validation History
•Looman and Abracen (2010) found the RRASOR generated a moderate AUC score of .62 for sexual recidivism.
b) International Research
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
None available at present.
•Parent, Guay and Knight (2011) the RRASOR achieved moderate accuracy in predicting sexual recidivism in a sample of 590 individuals who had committed sexual offences (AUC =.70).
No empirical evidence at present.
No empirical evidence at present.
b)International Research
Validation History
a)UK Research
Applicability: Mental Disorders
b)International Research
•Sjöstedt and Långström (2002) in a sample of individuals diagnosed with personality disorder who had committed sexual offences, the RRASOR demonstrated good predictive accuracy in relation to sexual recidivism (.73) and moderate accuracy in violent non sexual recidivism (.62).
•Blacker et al. (2011) the tool’s accuracy in predicting recidivism in a sample of individuals with learning disabilities who had committed sexual offences was below chance (AUC = .47).
•The RRASOR can provide a brief scan of the risk of sexual recidivism. It is an actuarial scale which can create awareness of some static risk factors related to the individual’s risk of sexual reoffending.
Contribution to Risk Practice
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
•Craig, Browne and Stringer (2004) RRASOR was able to distinguish between individuals residing in community settings to those in regional secure psychiatric settings with higher mean scores observed for the latter group.
•Långström (2004) the RRASOR attained good AUCs for individuals of Nordic (.76) and European (.77) ethnic origin for sexual recidivism. Despite this, it was unable to significantly predict recidivism in individuals of African Asian origin (.48).
Other Considerations
Applicability: Females
Applicability: Ethnic Minorities
Validation History
RATED page updated: August 2019 © Risk Management Authority 2019
•The authors recommend the use of the Static 2002 over the use of the RRASOR as the aforementioned tool includes more items, has been extensively cross validated and has a higher predictive accuracy than the RRASOR (see www.static99.org for more information).
•The tool may be of limited use with first time offenders whose current offence may be unusual or contain sadistic elements (Beech, Fisher and Thornton, 2003; Craig, Browne and Stringer, 2004).
•The RRASOR is normed on adult males with a previous offence history and is deemed inappropriate for the use with females, juveniles and individuals who are mentally ill.
•The author advises that the RRASOR should only be used to screen individuals who have committed sexual offences.
Assessor Qualifications
•Identifies factors for treatment
•The instrument does not employ actuarial or statistical methods to support decision making about risk, but instead offers guidelines for collecting relevant information and making structured decisions. The manual recommends that identified scenarios should discuss the nature, severity, imminence and likelihood of future sexual violence.
Strengths
Description
RATED page updated: August 2019 © Risk Management Authority 2019
•It is aimed at evaluating men aged 18 and over and may also be used with older male adolescents aged 16 and 17 and adult women with a degree of caution. It is not to be used with children aged 15 and younger (Hart and Boer, 2010).
Age Appropriateness
The manual prescribes that training may be completed via self study or attending lectures and workshops. The authors recommend 16 to 32 hours of training covering the following areas: Knowledge of sexual violence; Expertise in individual assessment; Expertise in mental disorder (Hart and Boer, 2010).
18+
•Provides a checklist for ensuring relevant factors have been considered.
•The RSVP produces a structured professional judgement assessment which has been guided by psychological theory. It is intended to help evaluators conduct comprehensive assessments of risk of sexual violence in clinical and forensic settings.
Name of Tool Risk for Sexual Violence Protocol (RSVP)
•The primary intended use of the instrument is to allow for forward planning in individual cases, guiding clinical decisions about risk assessment and management. A secondary use is ‘backward
•It can be used to identify the nature of risk for sexual violence and to develop and inform risk management strategies. It defines sexual violence as the “actual, attempted or threatened sexual contact with another person that is non consensual” (Hart et al., 2003).
•The RSVP is a 22 item structured guide for the assessment of those who have committed sexual offences, divided into five domains: sexual violence history, psychological adjustment, mental disorder, social adjustment and manageability.
Category Sexual Offending (Validated)
Year 2003
•Provides a means to measure the presence of risk factors
Author / Publisher Hart and Colleagues
•Hart, Michie and Cooke (2007) moderate predictive validity between the ‘Case Prioritisation’ scores and recidivism (r =.31) in a sample of adult males.
Validation History
RATED page updated: August 2019 © Risk Management Authority 2019
Inter Rater Reliability
•Watt and Jackson (2008) Excellent intra class correlation (ICC) obtained for ‘Presence Past Ratings’ (.95), ‘Present Recent Ratings’ (.85), ‘Case Prioritisation’ (.75) and ‘Risk of Harm’ (.81).
b) International Research
•Sutherland (2010) for steps 2 3 (‘Identification of risk item presence and relevance) and 6 (‘Summary Judgements’), the level of overall agreement was .53 (ICC2) amongst multi disciplinary forensic mental health clinicians.
•Watt and colleagues (2006) found similar ICCs within the .90s for ‘Presence Past Ratings’ (.96), ‘Present Recent Ratings’ (.96) and ‘Conclusory Opinions’ (.92).
•An unpublished doctoral thesis found the IRR of the RSVP was excellent for individual risk factors, summary risk ratings and total scores, ranging from .85 to .96 (Jackson, 2016).
•The RSVP can be used at various stages of the legal process from sentencing through to parole and in both inpatient and outpatient settings (Jackson, 2016).
looking evaluations’ to be used as a basis to evaluate the quality of risk assessments completed by other people (Hart and Boer, 2010).
The instrument was developed through literature review, revising guidance and feedback from users. It also underwent field testing of improvements in Canada and the UK (Hart and Boer, 2010).
a)UK Research
•Sutherland et al. (2012) found fair to moderate levels of agreement in relation to the summary judgement risk ratings and supervision recommendations in a sample of professionals within forensic mental health and learning disability settings. The highest inter rater reliability (IRR) was observed for professionals who were highly trained in forensic risk assessment. Other factors such as the complexity of the case and the number of training days attended for the RSVP also affected the IRR.
Empirical Grounding
General Predictive Accuracy
•The RSVP produces explicit guidelines for risk formulation centring on risk scenarios and management strategies (Hart and Boer, 2010).
a)UK Research
•Darjee et al. (2016) found that in a population that were likely to pose a higher risk of harm than a general sex offending population, they reported case prioritisation was significantly associated with time to sexual offending and time to breach but not time to violent offending. They also indicate that predictive validity is influenced by the level of case management. A decision on predictive validity for sexual offending or other offending is, therefore, unrealistic.
b)International Research
No empirical evidence at present.
Validation History
Applicability: Mental Disorders
Applicability: Females
There is a specific section in the evaluation that addresses mental disorder as relate to Diagnostic and Statistical Manual of Mental Disorders (DSM 5) or International Statistical Classifications of Diseases (ICD 10) diagnoses.
•The instrument showed good concurrent validity with the SVR 20, Static 2002R and the SORAG (Jackson, 2016).
Validation History
None available at present.
No empirical evidence at present.
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
Contribution to Risk Practice
•Darjee et al. (2016) argued that "tools such as the RSVP are good for identifying low risk individuals who do not require risk management".
Other Considerations
Applicability: Ethnic Minorities
•The RSVP is a derivative of the SVR 20 (Hart et al., 2003: 50), with a greater emphasis on psychological risk factors and developing case management plans. It is based on the sexual offending literature.
•Darjee et al. (2016) suggested that the RSVP may be a better tool for assessing the risk of serious harm in individuals who commit sexual offences rather than assessing their risk of sex offending. Findings of their study support the use of the instrument for the minority of those who pose a risk of serious harm. The authors related these findings to the Scottish criminal justice process, findings that the RSVP may be suitable for those being managed at MAPPA levels 2 and 3 and under consideration for an Order of Lifelong Restriction.
RATED page updated: August 2019 © Risk Management Authority 2019
•The instrument should not be used to determine whether someone who committed acts of sexual violence in the past nor to estimate the probability that sexual violence acts will be committed in future (Hart and Boer, 2010).
Author / Publisher Hanson, Harris, Scott and Helmus
Strengths
Name of Tool Stable 2007 and Acute 2007 (SA07)
Age Appropriateness
•The Stable 2007 measures ‘stable dynamic’ risk factors which are potentially changeable but may endure for months or years. The instrument incorporates a guided interview schedule which covers 13 major risk areas: significant social influences, capacity for relationship stability, emotional identification with children, hostility toward women, general social rejection, lack of concern for others, impulsivity, poor problem solving skills, negative emotionality, sex drive and preoccupation, sex as coping, deviant sexual preferences and cooperation with supervision. Items are scored on a 3 point scale of 0 to 2 from no problem, some problem and significant (Smid et al., 2014).
•The tools can be used to inform assessors with regard to level of priority and inform decisions on community treatment and supervision.
•The tools are easier to score than their predecessors, the Stable and Acute 2000 (Hanson et al., 2007).
•The Stable and Acute 2007 instruments have risk factors potentially aligned with pervasive developmental disorders: emotional identification with children, lack of concern for others, poor
Assessors must possess the relevant training and experience in conducting sexual violence risk assessments.
•These specialised tools aid the assessment of and track changes in an individual’s risk or motivation to change (Hanson et al., 2007).
Description
•The Acute 2007 measures ‘acute dynamic’ risk factors defined as highly transient conditions which can change over a period of weeks, days or even hours. The instrument assesses 7 areas of risk: victim access, hostility, sexual preoccupation, rejection of supervision, collapse of social support, emotional collapse and substance abuse. These items are scored on a 4 point scale from 0 for no problem, 1 for some problem, 2 for a significant problem and IN for intervene now (Smid et al., 2014).
•The Stable 2007 and Acute 2007 (SA07) is a two part actuarial risk assessment instrument designed to assist with the community supervision of individuals who have committed sexual offences.
18+ Assessor Qualifications
Category Sexual Offending (Validated)
RATED page updated: August 2019 © Risk Management Authority 2019
Year 2007
Empirical Grounding
Inter Rater Reliability
a)UK Research
a)UK Research
•Both instruments are grounded in the risk factors identified from the SONAR (Sex Offender Need Assessment Rating), STEP (Sex Offender Treatment Evaluation Project) Deviance (Beech et al., 2002) and Structured Risk Assessment (SRA) (Thornton, 2002), all of which were used in the prospective research design of the SA07.
•McNaughton Nicholls et al. (2010) both the Stable 2007 and Acute 2007 obtained poor to moderate ICCs ranging from .04 to .59 for all raters.
•Eher et al. (2010; 2011) found an excellent ICC value of .90 for the Stable 2007 composite score.
•The SA07 was based on the database compiled by the Dynamic Supervision Project (see http://www.publicsafety.gc.ca/res/cor/sum/cprs200709 eng.aspx for further details on this project).
No empirical evidence at present.
RATED page updated: August 2019 © Risk Management Authority 2019
•Fernandez (2008) found an overall ICC of .92 for Stable 2007. Individual item ICCs ranged between .56 and .91.
•Eher et al. (2010) found that the Stable 2007 was a strong predictor of sexual recidivism (AUC = .77) in a sample of Austrian child molesters.
problem solving skills, sex as a coping mechanism, capacity for relationship stability, negative emotionality and sexual preoccupations (Fabian, 2011: 77).
Validation History
b)International Research
b)International Research
•The Stable 2007 may not be entirely appropriate for those who have committed paedophilic sex offences. ROC
•Hanson et al. (2007) ICCs for the individual Acute 2007 items ranged from .64 to .95, with a median of .90; although no ICC was reported for the composite score.
General Predictive Accuracy
•Eher et al. (2011) the Stable 2007 attained moderate AUC values in predicting sexual reoffence (.71), violent reoffence (.67) and custodial sentence following violent offence (.69).
•Hanson et al. (2007) using 3 items from the Stable 2007, the authors found low to moderate ROC values for sexual (.52 .68) and violent (.51 .59) recidivism and any criminal offence (.50 .61).
Applicability: Ethnic Minorities
Validation History
No empirical evidence at present.
Validation History
Applicability: Mental Disorders
•Tamatea (2014) applied the STABLE 2007 to 245 males in New Zealand, finding that an AUC of 0.78 was yielded for reimprisonment. Over time, the STABLE 2007 was found to discriminate between higher and lower rate offending, albeit not for individuals with mid range scores.
•In a Canadian sample of 180 individuals convicted of sexual offences, it was found that the STABLE 2007 pre and post treatment scores were associated with sexual, non sexual violent and any violent recidivism (Sowden and Olver, 2017).
RATED page updated: August 2019 © Risk Management Authority 2019
•Hanson, Helmus and Harris (2015) found the STABLE 2007 scores added incrementally over STATIC scores in a sample of 768 Canadian individuals for all recidivism outcomes, but only for complete cases.
Applicability: Females
analyses and Cox Regressions were carried out on 189 prison released individuals who had committed paedophilic sex offences using a variety of instruments. It was found that the VRS:SO predicted sexual recidivism in this sample significantly better than the Stable 2007 and the Stable 2007/Static 99 combined score (Eher et al., 2015).
No empirical evidence at present.
Validation History
•Interviews with 24 probations service staff members in Ireland found that SA07 is perceived to be a practical risk assessment, directing supervision work and interventions (Walker and O’Rourke, 2013).
RATED page updated: August 2019 © Risk Management Authority 2019
•The SA07 is useful in identifying risk and responsivity factors pertinent to the individual’s risk of sexual recidivism.
Other Considerations
•The SA07 can contribute to an awareness of risk factors that may contribute to offending behaviour, examining stable dynamic risk factors that can be changed through treatment or supervision such as learned behaviours and personal skills deficits.
No empirical evidence at present.
•New revisions of the Stable and Acute 2007 manuals were published in 2012 with comprehensive Acute scoring guidance planned by the authors.
•The tool can inform the levels of monitoring and rehabilitation efforts needed to manage the case.
•The tool is used in the UK, Ireland, Canada, Germany and the United States for individuals convicted of sexual offences who are in community and prison settings.
•Assessors should note that the SA07 have been designed to aid the supervision of individuals who have committed sexual offences within community settings.
•The implementation of the SA07 was evaluated in Scotland. Please refer to the RMA website for updates (https://www.rma.scot/)
Contribution to Risk Practice
•No option to omit items within both tools in the instance where there is little to no information to score items.
•Inmaterials)…”theirmeta analysis of recidivism rates of females, Cortoni, Hanson and Coache (2010) concluded that risk assessment tools developed specifically for males who have committed sexual offences would be expected to substantially overestimate the recidivism risk of sexual offending.
•The Risk Management Authority (2013) has carried out research into practitioners involved in the implementation of SA07 in Scotland. This report highlighted potential learning points and support for training.
•Previous concerns regarding the ‘Access to Victims’ item (Mann, Hanson and Thornton, 2010; McNaughton Nicholls et al., 2010: 18) have been addressed in a draft Acute 2007 manual in which the scoring makes a distinction between chance events and deliberate actions that would otherwise increase the individual’s likelihood of recidivism (e.g. victim moving to house next door versus the perpetrator deliberately engaging in behaviours that would encourage the contact between themselves and potential victims). Fernandez and colleagues (2012: 19) advise that the SA07 is unsuitable for individuals whose only sexual offences are Category ‘B’ offences. These include offences where the “…participants were consenting (e.g., prostitution), the offence lacked a sexual motive (e.g., urinating in public), or there was no identifiable victim (e.g. possession of indecent
•Limited validation research on the Acute 2007 tool.
•The total risk score is used to classify individuals who have committed sexual offences into nine risk categories known as ‘bins’ (Rossegger et al., 2013).
Assessor Qualifications
Expertise in risk assessment for sexual offending. Assessors must also undergo the relevant training on the instrument.
Category Sexual Offending (Validated)
Name of Tool Sex Offender Risk Appraisal Guide (SORAG)
•The SORAG is a 14 item actuarial scale designed principally to assess risk for violent recidivism (including sexually violent recidivism) among adults released into the community. Items on the scale are: living with biological parents until age 16; elementary school maladjustment; history of alcohol problems; never been married at time of index offence; criminal history scores for nonviolent and violent offences; number of convictions for previous sexual offences; history of sexual offences (for girls under the age of 14); failure on prior conditional release; age at index offence; diagnosis of schizophrenia or any other personality disorder; phallometric test; psychopathy checklist (Rice and Harris, 2016).
Description
Empirical Grounding
Age Appropriateness
Author / Publisher Quinsey, Harris, Rice and Cormier
•The PCL:R (Hare, 2003) score features as an item within the SORAG. It uses clinical records as a basis for scoring and incorporates PCL:R scores.
Strengths
•The SORAG is an extension of the Violence Risk Appraisal Guide (VRAG) and shares ten items with it (Parent, Guay and Knight, 2011; Quinsey et al., 2006).
•The SORAG generally appears to exhibit higher correlations with violent recidivism in comparison to other instruments (e.g. Static 99 and SVR 20) developed for use with individuals who have committed sexual offences. (Rettenberger and Eher, 2007).
18+
Year 1998
•The intended and recommended purpose of the SORAG is to render an estimate of the long term risk of criminal violence in general.
RATED page updated: August 2019 © Risk Management Authority 2019
a)UK Research
•Ducro and Pham (2006) excellent ICC of .92 attained for the SORAG.
General Predictive Accuracy
•Rettenberger and Eher (2007) the SORAG achieved an ICC value of .93 in a German sample of individuals committed of sexual offences.
•Langton et al. (2007) large correlation coefficient of .90 observed for inter rater reliability of the SORAG.
•Parent, Guay and Knight (2011) found the SORAG had a moderate AUC of .69 for predicting recidivism in a group of 590 individuals who had committed sexual offences.
Inter Rater Reliability
a)UK Research
None available at present.
•Rice and Harris (2016) found the SORAG yielded high predictive accuracy for general and violent recidivism, with the AUC giving a score of 0.73.
•Rettenberger et al. (2010) the SORAG generated an AUC value of .68 for general violent recidivism and .72 for general criminal recidivism within a subgroup of rapists.
None available at present.
•Rettenberger et al. (2017) examined the German version of the SORAG in a sample of 1104 individuals in Austria. The SORAG was found to have a small but significant predictive validity over the VRAG and PCL R, yielding a moderate AUC of 0.74.
b)International Research
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
The SORAG draws on the empirical literature in relation to sexual recidivism and that related to the VRAG. Ten of the items are derived from the VRAG (Rettenberger and Eher, 2007)
•Walters, Knight and Thornton (2009) the SORAG generated a high ICC score of .89
•A study of 137 individuals in Switzerland gave moderate AUC scores of 0.69 and 0.67 for total risk scores and risk bins respectively (Rossegger et al., 2013).
b)International Research
Validation History
•Looman (2006) moderate AUC scores attained for the SORAG (.69).
No empirical evidence at present.
Within the extra familial molestation subgroup, the SORAG exhibited moderate to high accuracy in predicting sexual recidivism (.71), sexual violent recidivism (.62), general violent recidivism (.81) and general criminal recidivism (.77).
•Eher et al. (2008) the SORAG was found to be a highly predictive accurate tool, particularly with a subgroup of individuals who had committed child sexual offences (AUC = .82).
•Langton et al. (2007) the SORAG demonstrated moderate accuracy in predicting serious violent offending (.71).
b)International Research
Applicability: Mental Disorders
Applicability: Females
a)UK Research
Validation History
•Pham and Ducro (2008) found moderate AUCs for prediction of general recidivism (.69), violent recidivism (.71) and sexual recidivism (.62).
No empirical evidence at present.
•The predictive validity of the SORAG was tested in 258 adult males. Sexual recidivism yielded an AUC of .65; this was in spite of a relatively low risk in the sample, given 53.5% had committed incest offences. The AUC generated for violent recidivism was .69 (Nunes et al., 2002).
Validation History
RATED page updated: August 2019 © Risk Management Authority 2019
Applicability: Ethnic Minorities
None available at present.
•Ducro and Pham (2006) the SORAG achieved moderate AUC values ranging from .64 to .65 in a sample of those convicted of child abuse and rape from a high secure hospital.
•The SORAG is a risk assessment tool comprised solely of static variables and therefore it is not possible to select treatment targets, measure change or progress in treatment or predict the time frame in which an individual is likely to re offend (Yates, 2005).
RATED page updated: August 2019 © Risk Management Authority 2019
Contribution to Risk Practice
Other Considerations
•The SORAG provides a brief assessment of the risk of sexual recidivism and can prompt further analysis of the identified risk.
•The tool is time consuming to administer and is more difficult to score as some of the items are taken from the VRAG.
•The tool relies on PCL:R rating scores as part of the predictive measurement.
• It has been found that the SORAG has better accuracy in predicting violent rather than sexual recidivism (see Rettenberger and Eher, 2007).
•Other investigations have found that this instrument has better predictive accuracy with different sub groups of individuals who committed sexual offences (Ducro and Pham, 2006).
•The SORAG shows some consideration for responsivity issues (e.g. psychopathy).
•The authors of the instrument recommend that the revised version of the scale (Static 2002R) replace the Static 2002 and the Static 99/Static 99R in all contexts where it is used (Phenix et al., 2009).
•The convergent validity of the Static 99R with the Static 2002R was found to be almost perfect, with a mean Cohen’s Kappa of .86 (standard deviation=.18) (Brouillette Alarie, Prolux and Hanson, 2017).
•The items presented in this measure are identical to those in the Static 2002 with the exception of the updated age weights (see Helmus et al., 2012). The revision was to improve consistency across scoring categories, conceptual clarity and predictive accuracy. These items are grouped into five categories: age at release, persistence of sex offending, sexual deviance, relationship to victims and general criminality (Brouillette Alarie, Prolux and Hanson, 2017).
Assessor Qualifications
Strengths
Category Sexual Offending (Validated)
Year 2012
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
18+
This tool is derived from the Static 99 and is grounded in research literature relating to sexual offending, as well as specific literature regarding the Static 99 (Helmus, 2009).
Name of Tool Static 2002R
Description
•Can be used by a variety of professionals such as psychologists, police officers and probation officers.
Experience in assessing sexual violence risk. Training on the instrument is highly recommended. Note that Static 99/R training is not sufficient to score Static 2002/R. The authors recommend that evaluators obtain Static 2002/R training before using the scale. http://www.static99.org
Empirical Grounding
•It is a 14 item actuarial risk measure designed to predict sexual recidivism in adult males who have committed sex offences.
•Individuals can be placed into one of five categories based on their total score which range from low, low moderate, moderate, moderate high to high risk (Babchishin, Hanson and Helmus, 2012).
Author / Publisher Helmus and Colleagues
a)UK Research
Validation History
b)International Research
General Predictive Accuracy
•In a sample of 342 individuals who had committed sexual offences, ROC analyses demonstrated an AUC of .769 (Jung et al., 2017).
None available at present.
•Helmus et al. (2012) observed a slight increase in the predictive accuracy of the Static 2002R compared to the Static 2002 for sexual recidivism at the 5 year follow up period (AUCs = .713 and .709 respectively). For violent recidivism, however, there were no observed differences in the AUC values. None of the differences between the two scales were significant.
Validation History
•Babchishin, Hanson and Helmus (2012) the Static 2002R demonstrated accuracy in predicting sexual recidivism (AUC =.76).
•Babchishin, Hanson and Helmus (2011) the Static 2002R showed moderate accuracy in predicting sexual recidivism (AUC =.70).
•The AUC of the Static 2002R was .69 in a sample of 590 individuals who had committed sexual offences (Parent, Guay and Knight, 2011).
b)International Research
None available at present.
•The AUC scores for 5, 10 and 15 year follow ups in a sample of 621 Australian individuals convicted of sexual offences were .68, .67 and .69 (Reeves, Ogloff and Simmons, 2017).
•Jung et al. (2017) found that 4 of the Static 2002R items demonstrated relatively poor IRR limits; although this may be due to shortcomings in training for three out of the four.
Inter Rater Reliability
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
•In his doctoral thesis, Lee (2019) conducted fixed effect meta analyses from five independent Canadian samples to test the predictive accuracy of the Static 2002R with White (n=1560) and Indigenous (n=653) groups. The Static 2002R was able to discriminate recidivists from non recidivists in the group of White individuals (AUCs>.69). For the Indigenous group, however, the predictive accuracy was not statistically significant (AUC<.61). Consequently, extreme caution is urged when using Static 2002R with individuals of Indigenous heritage.
•Limited studies conducted by independent researchers.
b)International Research
Validation History
•Assessors are encouraged to use the BARR 2002R over the Static 2002R or the Static 99 when predicting violent or any recidivism (Babchishin 2019, personal communication).
•The Static 2002R provides a brief scan of the risk of sexual recidivism and can prompt further assessment of identified risk factors.
RATED page updated: August 2019 © Risk Management Authority 2019
None available at present.
•Babchishin, Hanson and Helmus (2012) Aboriginal individuals had significantly higher composite scores than non Aboriginals. The Static 2002R composite score did not significantly predict sexual recidivism for Aboriginal individuals.
No empirical evidence available.
Applicability: Females
a)UK Research
Validation History
Applicability: Ethnic Minorities
Applicability: Mental Disorders
Other Considerations
Contribution to Risk Practice
Not intended for use with females.
•Hanson and colleagues (2017) created five new risk categories for the Static instruments and found that this increased the concordance of risk classification from 51% to 72. It is theorised that the new common STATIC risk categories could inform intervention strategies.
•It is recommended that those in the community who are offence free for five years should have their risk category reduced by one level. The term ‘offence free’ is interpreted as no known criminal activity taking place, no convictions, no warnings, reprimands or breaches (Hanson et al., 2014; Hanson et al. (2018).
RATED page updated: August 2019 © Risk Management Authority 2019
•For further information, please visit the following website; www.static99.org
•In a Master’s thesis, Rohrer (2019) carried out factor analyses with a sample of individuals who had committed sexual offences (n=533). A potential new factor structure was proposed for the Static instruments: paedophilia factor, young antisociality factor, general criminality factor and agonistic continuum factor.
Author / Publisher Boer, Hart, Kropp and Webster
Year 2018
•The SVR 20 is useful in assisting the structuring of clinical assessments and also incorporates a ‘recent change’ score.
•SVR 20 is a 20 item structured framework published in 1998 intended to evaluate risk of sexual violence and informing risk management. A second version was published in 2018
Empirical Grounding
Assessors must possess the relevant training and experience in conducting sexual violence risk assessments. Certified workshops are available through the Global Institute of Forensic Research: https://www.gifrinc.com/course/svr 20 demand training/
The SVR 20 was developed from a thorough consideration of the empirical literature concerning factors that relate to sexual violence. The twenty items found in the SVR 20 have appeared to exhibit a relevant association with sexual recidivism and appear to be relevant in regard to clinician experience (Knight and Thornton, 2007). The SVR 20 manual serves as an organised literature
Assessor Qualifications
RATED page updated: August 2019 © Risk Management Authority 2019
Description
18+
Strengths
Category Sexual Offending (Validated)
•The 20 items are organised under three subsections: (1) Psychosocial Adjustment, (2) Sexual Offenses, and (3) Future Plans. The items covered in each subsection are: 1) psychological adjustment, sexual deviance, victim of child abuse, cognitive impairment, suicidal/homicidal ideation, relationship/employment problems, previous offence history (non sexual violent, non violent), psychopathy substance use problems and past supervision failure; 2) sexual offending high density offences, multiple offences, physical harm to victims, use of weapons, escalation and cognitive distortions; 3) Future plans lacks future (realistic plans) and has negative attitudes towards instruction. The items are coded as absent, possibly or partially present and present (Hart and Boer, 2010).
•The SVR 20 is based on structured clinical judgment and was developed for use with forensic mental health populations (Parent, Guay and Knight, 2011).
Age Appropriateness
Name of Tool Sexual Violence Risk 20 Version 2 (SVR 20 V2)
•Ramirez et al. (2008) ROC curve analysis demonstrated discriminate capacity for the SVR 20 with an AUC value of .83.
•Rettenberger et al. (2010) the SVR 20 scores were found to be highly associated with most types of recidivism for a subgroup of Extra Familial Molest Offenders for a range of offences: sexual recidivism (r =.75), sexual violent recidivism = .51, general violent recidivism = .81 and general criminal recidivism = .77.
b)International Research
Inter Rater Reliability
•de Vogel et al. (2004) large ICC value of .75 observed for the SVR 20
a)UK Research
Validation History
a)UK Research
b)International Research
•Hill et al. (2008) higher scores on the SVR 20 predicted higher recidivism rates within the sample of non sexually violent (previous convictions) offenders.
•Sjöstedt and Långström (2002) Kappa values ranged from .60 and ICCs were also moderate (.62).
review for sexual violence risk, simply listing research supporting each item found on this assessment (Witte, 2001).
RATED page updated: August 2019 © Risk Management Authority 2019
•Parent, Guay and Knight (2011) found the SVR 20 yielded a moderate AUC of .66 for sexual recidivism in a sample of 590 sex offenders.
•The SVR 20 was tested on 493 sex offenders from the Austrian prison system. It showed good predictive accuracy for the prediction of sexual recidivism for the entire sample (AUC=.72); although some inconsistencies were present based on recidivism criterion and offender subgroup (Rettenberger, Boer and Eher, 2011).
None available at present.
None available at present.
•Knight and Thornton (2007) ICCs were below .70.
General Predictive Accuracy
Applicability: Mental Disorders
•Dietiker, Dittmann and Graf (2007) the SVR 20 was found to have good predictive capacity within hospital settings (AUC = .88).
•Knight and Thornton (2007) at the 3 , 10 and 15 year follow up, the SVR 20 demonstrated moderate AUC values for serious sexual charges (.66, .68, and .68 respectively).
a)UK Research
Not intended for use with female offenders
•Craig et al. (2006a) the SVR 20 was only able to moderately predict offence reconviction for ‘any offence’ at the 5 and 10 year follow up points in a sample of offenders referred to a regional secure unit (.61 and .60 respectively). It did not, however, significantly predict sexual and violent recidivism.
b)International Research
Validation History
Applicability: Females
Validation History
•Kanters and colleagues (2017) tested the SVR 20 on 639 sex offenders in a forensic psychiatrist centre. The pre treatment SPJ score significantly predicted general recidivism (AUC=.71); whilst the post treatment score significantly predicted sexual, violent and general recidivism with AUCs of .76, .75 and .70 respectively.
Validation History
No empirical evidence at present.
Applicability: Ethnic Minorities
•Craig et al. (2006b) the SVR 20 scores had small correlations with general recidivism in the follow up 2 , 5 and 10 year periods which ranged between .18 to .24.
RATED page updated: August 2019 © Risk Management Authority 2019
•Sjöstedt and Långström (2002) the SVR 20 did not significantly predict recidivism in a sample of personality disordered offenders.
RATED page updated: August 2019 © Risk Management Authority 2019
•The SVR 20 can aid assessors in identifying risk and responsivity factors specific to the individual (e.g. criminal lifestyle, presence of mental health problems). The factors identified can also act as targets for change.
Contribution to Risk Practice
•Blacker et al. (2011) found the composite score achieved low to high AUC values in predicting sexual (.45), violent (.80) and general (.50) recidivism in a sample of offenders with learning disabilities.
•The SVR 20 can aid detailed assessment of the risk of sexual recidivism.
•The tool can contribute to the formulation of offence analyses and risk management strategies.
•The instrument itself is normed on adult males with a previous offence history and is deemed inappropriate for use with females or juveniles.
•SVR 20 V2 is available through its publisher: https://www.parinc.com/Products/Pkey/4534
Other Considerations
Author / Publisher Wong, Olver. Nicholaichuk and Gordon
Category Sexual Offending (Validated)
•Olver et al. (2016) found that the VRS SO test scores demonstrated construct validity risk in those who committed sexual offences.
Strengths
•The measure is comprised of 7 static and 17 dynamic items which are empirically, theoretically or conceptually linked to sexual recidivism. It is measured on a 4 point scale from 0 3.
Assessors must possess the relevant training and experience in conducting sexual violence risk assessments. Assessors should also undergo the relevant training for this tool.
•The tool is deemed advantageous in regard to assessing risk and identifying an individual’s motivation for change based on dynamic risk factors over the course of a period of time or treatment (Maltais and Sribney, 2018).
Name of Tool Violence Risk Scale: Sexual Offenders (VRS:SO)
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
Assessor Qualifications
Year 2003 Description
•The sexual deviance, criminality and treatment responsivity factors of the VRS:SO were found to correlate in significantly meaningful ways with the Stable 2000; thus, indicating its psychological constructs relate to risks and needs in terms of sexual offending (Olver et al., 2018b).
Empirical Grounding
18+
•The VRS:SO is a 24 item assessment derived from the original Violence Risk Scale (VRS).
•The VRS:SO is designed to assess risk of sexual recidivism in forensic populations.
•The CPORT risk tool and CASIC scale (Seto and Eke, 2015) found that there were moderate positive correlations between the VRS:SO’s criminality and sexual deviance scores respectively. this indicates an underlying construct for measuring risk in those who have committed only internet offences (Maltais and Sribney, 2018).
•The measure generates pre and post treatment composite scores related to therapeutic change and risk change.
•The tool is designed to measure change in the level of risk before and after treatment/intervention. The VRS:SO scores are used to inform case conceptualisation and treatment planning (Olver et al., 2018b).
•Sowden (2013) pre and post treatment scores on the VRS:SO were found to have moderate accuracy in predicting sexual (AUCs = .61 and .62, respectively) and violent (AUC = .63 and .66, respectively) recidivism in a sample of high risk Canadian males who had received treatment for their sexual offending.
•Eher and colleagues (2015) administered a number of risk assessment tools to a sample of paedophilic individuals and carried out ROC and Cox Regression analyses to test predictive accuracy. The VRS:SO significantly predict sexual recidivism, more so than the PCL:R, Stable 2007 and Static 99/Stable 2007 combined score. Moreover, it was found that when this was combined with an exclusive diagnosis of paedophilia incremental validity was added.
General Predictive Accuracy
•Sowden (2013) reported ICCs of .90 and .86 for pre treatment and post treatment total scores in a randomly selected subsample of treated high risk Canadian adult males.
b)International Research
•Olver et al. (2012) pre and post treatment scores were found to significantly predict sexual and violent recidivism (AUCs = .65 and .67 for pre and post respectively across
RATED page updated: August 2019 © Risk Management Authority 2019
•Beggs and Grace (2010) found excellent ICCs of .90 and .92 for the pre and post treatment dynamic total scores on the VRS:SO.
The Manual states that each of the static and dynamic items found in the tool are grounded in empirical research drawn from risk assessment literature, with theoretical underpinnings from Andrews and Bonta’s (2010) Psychology of Criminal Conduct, advances in relapse prevention theory, and the ‘Stages of Change’ model (Olver et al., 2014; Prochaska and DiClimente, 1992)
a)UK Research
None available at present.
None available at present.
a)UK Research
Validation History
•Olver et al. (2007) the VRS:SO also achieved high ICC values for the composite pre and post treatment scores (.74 and .79 respectively).
b)International Research
Inter Rater Reliability
RATED page updated: August 2019 © Risk Management Authority 2019
•Sowden and Olver (2017) assessed a sample of 180 Canadian individuals who had committed sexual offences. VRS: SO scores were found to predict sexual, non sexual violence, any violence (including sexual) and general recidivism.
both outcomes) in a prospective multisite Canadian study of 571 individuals who had committed sexual offences.
•Olver et al. (2007) the composite pre and post treatment scores generated AUCs of .71 and .72 respectively.
•A study by Olver and colleagues (2016) of those who had committed sexual offences (n=668) found the VRS:SO items and total scores predicted sexual, violent and general recidivism for five and ten year follow up periods.
•Beggs and Grace (2010) found large AUCs of .78 for pre and .81 for post treatment scores in relation to sexual recidivism.
Validation History
Applicability: Females
No empirical evidence available.
Applicability: Ethnic Minorities
•In a sample of 539 individuals convicted of sexual offences in New Zealand and Canada, Olver et al. (2014) found there was moderate to high predictive accuracy for a follow up average of fifteen and a half years.
•Sowden and Olver (2017) administered the VRS:SO to a Canadian sample of individuals who had committed sexual offences (n=180). Significant predictive accuracy was demonstrated for various types of recidivism: sexual, non sexual violent, any violent and general. A reduction in all types of recidivism was evident in VRS:SO change scores.
Validation History
No empirical evidence available.
RATED page updated: August 2019 © Risk Management Authority 2019
•Olver (2004) maintained "The use of logistic regression demonstrated a clinically useful and systematic means of combining risk and change information into post treatment risk appraisals."
Validation History
No empirical evidence available.
•The tool can contribute towards the measurement of progress or deterioration in factors related to the individual’s level of risk.
•The tool can aid assessors in the development of offence analyses and risk management strategies.
•Some VRS:SO items (e.g. community support, release to high risk situations) specifically address risk management responses of the individual in the community.
•Based on an initial investigation conducted by Maltais and Sribney (2018), there is evidence to suggest that the VRS:SO could be used with those who have only committed internet offences; although caution should be exercised until further research is conducted.
•Many of the factors included in the VRS:SO can be used to identify treatment needs and for treatment planning.
•The risk categories and recidivism estimates for the VRS:SO were recently updated by Olver and colleagues (2018a)
Other Considerations
Applicability: Mental Disorders
•The VRS:SO can enable assessors to identify static and dynamic factors relevant to the risk of sexual reoffending.
Contribution to Risk Practice
•The instrument is composed of a subset of six items from the Static 2002R (measure of general criminality) and age at release.
Tool Development
Year 2013
The18+Barr
Name of Tool Brief Assessment for Recidivism Risk (BARR 2002R)
2002R can be used with caution on individuals who committed their most recent offence between the ages of seventeen and eighteen on the condition that their release date falls after their eighteenth birthday. It is not to be used with those whose offence(s) all took place when they were under the age of seventeen.
Experience in assessing sexual violence risk. Training on the instrument is highly recommended.
Author / Publisher Babchishin, Hanson and Blais
•The authors recommend that the Barr 2002R is used to predict violent and general recidivism rather than the Static 2002R (Babchishin, Hanson and Blais, 2013).
•In the development study, the Barr 2002R was found to predict general and violent recidivism as well as other measures like the LS/CMI (Babchishin, Hanson and Blais, 2013).
Assessor Qualifications
•Jung and Wielinga (2019) suggested that the Barr 2002R may be a valuable tool for law enforcement to evaluate violence risk among individuals charged with sexual assault offences. Applying the Barr 2002R to individuals charged with sexual assault (n=293) found that the tool showed a large effect in its ability to predict future general and violent offending; although the severity of future violent offending was not associated with Barr 2002R scores.
•The Barr 2002R is an actuarial risk scale used to predict violent and general recidivism risk.
Category Sexual Offending (Awaiting Validation)
RATED page updated: August 2019 © Risk Management Authority 2019
Age Appropriateness
Description
•Jung, Wielinga and Ennis (2018) administered the Barr 2002R on males who had committed sexual offences (n=324). The instrument showed large effect sizes for predicting general and violent recidivism (AUC .718 and .737 respectively) and a moderate one for sexual recidivism predictions (AUC .661).
•The Barr 2002R was developed within a sample of individuals convicted of sexual offences; thus, cannot be used on those who have committed non sexual offences (Babchishin, Hanson and Blais, 2013).
General Notes
RATED page updated: August 2019 © Risk Management Authority 2019
•Babchishin, Hanson and Blais (2016) recommended that individuals scoring highly on the Barr 2002R (five or higher) should be subject to a more detailed risk assessment.
•Jung, Wielinga and Ennis (2018) found that there was a significant correlation between the scores of Barr 2002R and the SORAG.
•The Barr 2002R is not recommended for use in cases of sexual crimes like statutory rape where the sexual activity was consensual between people of similar ages. It is also not recommended for use in individuals who have only been convicted of offences such as child pornography possession and prostitution. It is also not to be used on those who have not committed a violent offence after living in the community for a period of at least eight years (Babchishin, Hanson and Blais, 2013).
•The CPORT is a risk assessment tool designed to predict any sexual recidivism for men convicted of child pornography offences.
Name of Tool Child Pornography Offender Risk Tool (CPORT)
Description
•It is recommended that the CASIC scale developed by the authors to assess paedohebephilia (sexual interest in children) is used in conjunction with the CPORT tool.
Assessor Qualifications
Tool Development
RATED page updated: August 2019 © Risk Management Authority 2019
Year 2015
Category Sexual Offending (Awaiting Validation)
•It consists of 7 items: age at time of investigation, any prior criminal history, any contact sexual offending, any failure on conditional release, indicators of sexual interest in child pornography material or prepubescent or pubescent children, more boy than girl content in child pornography and more boy than girl content in other child depictions. If present, each item is worth one point.
•It was developed from a sample of 286 adult men convicted of a least one child pornography offences using information available in Canadian police files, meaning that other criminogenic factors could not be examined. The legal definition for child pornography in Canada was used, where images of nude or partially dressed children are not illegal if there is no sexual activity and/or focus on the genital and anal regions (Eke and Seto, 2016).
The CPORT guide details the following requirements for those intending to use the tool: “Individuals accessing this document are expected to be experienced and current in the field of risk assessment and sexual offending, including child pornography offending, in order to make informed decisions regarding how the information herein may be relevant to their professional practice” (Eke, Helmus and Seto, 2018).
Age Appropriateness
Author / Publisher Seto and Eke
•The development samples were amalgamated with a new sample of 80 individuals with internet offences to give an overall sample of 346 men with internet offences. Any sexual recidivism for the
•The development study found the CPORT had acceptable predictive accuracy for any sexual recidivism (AUC=.74) and contact sexual recidivism (AUC=.74). The CPORT was found to significantly predict sexual recidivism for internet offenders with other types of offending (bar contact offences) (AUC=.69) and those with histories of contact offending (AUC=.80); although it did not significantly predict sexual recidivism for those with only internet offences (Seto and Eke, 2015).
18+
•At the present time, there has been no empirical assessment of the use of CPORT with individuals who have been charged and are still involved in criminal proceedings. If an assessor decides to use
entire sample was an acceptable level at .724. This sample was then divided by those who committed only internet offences and those who had also committed contact offences. Predictive accuracy was found to be moderate at .685 for those who had committed internet offences and .767 for individuals with contact and internet offences (Eke, Helmus and Seto, 2018).
•The authors advise evaluators use caution when reporting CPORT scores, given the unknown stability of the recidivism estimates. It is also recommended that the tool is not used if more than one item is missing (Eke and Seto, 2016).
•Further updates on the CPORT, including a regularly revised ‘Frequently Asked Questions’ section, are available at: https://www.researchgate.net/project/Child Pornography Offender Risk Tool CPORT
•A sample of 119 individuals convicted of child pornography offences were scored with the CPORT. CPORT scores are significantly higher for offenders with contact sexual offences and although the likelihood of child pornography reoffending is low, the odds increase with higher scores on the CPORT (Gunnarsdóttir, 2019).
RATED page updated: August 2019 © Risk Management Authority 2019
•As indicated in personal communications in 2017, Seto and Eke gauged there is interest in validating the CPORT in the United States, Australia and criminal justice agencies throughout Canada. In another email update in 2017, Eke shared that there are a few groups currently conducting or planning to conduct validation work with the instrument: for instance, Ontario’s probation and parole.
•The CPORT can be used with any man who has committed an internet offence. This includes men who have committed internet offences alongside other types of sexual offences (e.g. contact offence against a child). It can be used alongside other risk assessment tools meant to assess sexual recidivism.
•A modified version of the CPORT (CPORT M) was developed, consisting of five items: age, prior criminal history, any contact sexual offending, any failure on conditional release and any indication (and/or admission) or paedophilic or hebephiliac interests. The CPORT M was found to have moderate predictive accuracy of general recidivism amongst individuals with child pornography offences (AUC=.68); although the sexual recidivism rates were too low to correlate with risk (Pilon, 2016).
•A limitation of the tool is that the admission or diagnosis of paedophilia or hebephilia is vulnerable to self reporting bias. To negate this risk, Seto and Eke (2017) developed the CASIC scale to assess ‘paedophebephilia’ (sexual interest in prepubescent and pubescent children respectively) This scale consists of six items looking at factors like marital status, collection and nature of context, access to children and engaging in online communications with children. Testing the scale on the sample of males from the developmental study resulted in an acceptable AUC of .71. further testing on a small cross validation sample of 60 individuals with internet offences showed excellent predictive accuracy with an AUC of .81. The authors recommended that the CASIC measure may replace item 5 of the CPORT if a score of 3 or more is generated (Seto and Eke, 2017).
•A study looking at the convergent validity of CPORT with the VRS:SO found moderate positive correlations between the CPORT and VRS:SO criminality score. The CASIC scale was also found to have a moderate positive correlation with the VRS:SO sexual deviance score (Maltais and Sribney, 2018).
General Notes
•The CPORT had not been evaluated with females or juveniles; thus is not recommended for use in these populations (Eke and Seto, 2016).
RATED page updated: August 2019 © Risk Management Authority 2019
CPORT prior to conviction for the purposes of sharing information about risk factors, the authors highly recommend that a caveat is added in about there being no empirical support for this. The authors caution against using CPORT in cases where the individual’s charge for child pornography offences has already been withdrawn or dismissed there is currently no empirical evidence for using the CPORT in this population (Eke, Maaike Helmus and Seto, 2018).
•The CPORT is free and publicly available.
•The tool can be applied to males involved in possession, distribution and production of IIOC.
The KIRAT is intended for use within the police service. Assessors are required to complete an online training course. It is only suitable for use in adult males.
Description
Age Appropriateness
Author / Publisher McManus, Long and Alison
Category Sexual Offending (Awaiting Validation)
RATED page updated: August 2019 © Risk Management Authority 2019
•The tool contains summary risk judgements for the two pre assessment factors.
Tool Development
•The KIRAT 2 is a brief screening tool designed to assess the risk of contact sex offending in males who access indecent images of children (IIOC) via the internet. It is the national model in England and Wales to rank those who have committed internet offences for their risk of committing contact offences against children based on their history of contact offending. It aims to identify internet offenders who share the most features with those who have also committed contact sexual offences (Long et al,. 2016).
18+ Assessor Qualifications
•The factors included in the KIRAT are based on empirical literature relating to internet sex offending. The context for the creation of the KIRAT was related to the limited application of other validated risk assessment tools in assessing sexual violence risk posed by individuals who commit internet sex offences (Long, Alison and McManus, 2012; McManus et al., 2011).
Year 2011
•The KIRAT was generated subsequent to a collaborative project between the University of Liverpool and the Kent Police Force in relation to the investigation of the use of abusive and indecent images of children. The tool was created in order to aid police in prioritising cases of internet sex offending.
•The tool is comprised of two parts. Part one examines two pre assessment factors: (1) organisational risk and (2) risk of contact sexual abuse. Part two contains fourteen items that assess the individual’s previous offence history, their access to children and any other factors that may contribute to their level of risk.
Name of Tool Kent Internet Risk Assessment Tool 2 (KIRAT 2)
•The final risk level of committing contact sexual offences is categorised as either ‘low’, ‘medium’ or ‘high.’ The tool does not predict future risk or reoffending; rather, its intention is to provide a ‘robust procedure’ to prioritise cases (Long et al., 2016).
•An executive summary published by the Child Exploitation and Online Protection Centre (CEOP) described the KIRAT as "… currently the most rigorously tested and widely employed risk assessment tool for IIOC cases…" (CEOP, 2012).
•For further information, e mail: kirat@kent.pnn.police.uk
•Long and colleagues (2016) examined the validity of version 2 of the KIRAT. Data was obtained from 374 police files from 11 forces across the UK. The KIRAT classified 97.6% for high risk and 62.3% for low and medium risk. Fifty cases from the data set were randomly selected to test inter rater reliability. The majority of variables demonstrated excellent agreement with ICC values higher than .750 and the remainder showed good agreement. Long et al (2016) concluded that the KIRAT provides an evidence based approach to prioritising cases for law enforcement.
•The tool can be completed by hand or electronically.
RATED page updated: August 2019 © Risk Management Authority 2019
•The tool is widely used across police forces in England and Wales.
The development study looked at 273 individuals who had offended (120 with both internet and contact offences and 153 with non contact offences), with the KIRAT correctly identifying 100% of those at high risk and 70% of those who are low and medium risk of progressing onto a contact offence The predictive accuracy was found to be good with an AUC of .86. The following factors were identified as significant to dual offending: previous non sexual convictions, grooming, production of IIOC, living with partner and children who are not biologically their own and access to children (Long, Alison and McManus, 2012).
•The national roll out of the KIRAT commenced in 2012. The same year the European Commission provided funding under the ‘Fighting International Internet Paedophilia’ (FIIP) project with partners from the UK, Netherlands, Spain, Ireland and Estonia to develop the tool into its second version (Long et al., 2016).
General Notes
•The KIRAT is not a conventional risk assessment tool to be used for measuring recidivism. It is also not suitable to be used for sentencing purposes or assessment for court.
•Tool has not been validated with other specific groups such as females or adolescents
Age Appropriateness
Tool Development
18+
•The tool contains 16 items that assess risk, treatment needs and progress in those convicted of sexual offences. Each risk factor is scored either 0 (not present), 1 (present but not a central characteristic), 2 (a central characteristic). For instances when information is poor or inconsistent, it is not possible to score factors (Webster et al., 2006).
•The SARN is administered in conjunction with the RM2000. Utilising the RM2000 to assess the static factors, the SARN follows on thereafter to determine the dynamic and treatment factors.
•SARN can be completed by HPC Registered forensic psychologists and experienced probation officers.
•The SARN is based on the Structured Anchored Clinical Judgement Minimum (SACJ Min) which itself has been updated and refined into RM2000 (Thornton, 2002).
Description
Name of Tool Structured Assessment of Risk and Needs (SARN)
Assessor Qualifications
•The SARN is a standardised treatment planning tool used within the HM Prison Service (Hogue, 2009). It seeks to identify the long term psychological risk factors relevant to individuals who have committed sexual offences (Smid et al., 2014).
Assessors should have the relevant training offered by the HM Prison and Probation Service and experience in risk assessment and risk management and SARN can be completed by HCPC Registered forensic psychologists
•Webster et al. (2006) obtained a mean inter rater agreement of 84.3% in a pilot study with a small sample of trained professionals. In the four cases rated by the professionals, the Kappa values ranged from .61 to .84.
Author / Publisher Thornton Year 2002
•The tool is a derivative of the Structured Risk Assessment (Thornton, 2002).
RATED page updated: August 2019 © Risk Management Authority 2019
Category Sexual Offending (Awaiting Validation)
•The items are subdivided into four domains; (1) sexual interest: sexual preoccupation, sexual preference for children, sexualised violence preference and other offence related sexual interests; (2)distorted attitudes: adversial sexual entitlement beliefs, rape supportive beliefs, viewing women as deceitful; (3) social and emotional functioning: feelings of personal inadequacy, distorted intimacy balance, grievance thinking towards others, a lack of emotional intimacy with adults; (4) self management: impulsiveness, poor problem solving, poor management of emotion.
•Beech et al. (2003) reported on the use of the SARN in conjunction with the Sex Offender Treatment Programme in English prisons. While facilitators welcomed the tool, they found the outcome reports that were generated were too complex and lengthy for less literate individuals.
•Hocken and colleagues (2013) carried out thematic analysis on the transcript from a discussion group of sexual offending management experts. It was found that the SARN was not appropriate to be used with individuals with learning disabilities.
•Webster and colleagues (2006) carried out two studies: one with seven experienced raters; the other with clinicians with differing levels of experience who had been trained in the SARN. The first study demonstrated strong support for the inter rater reliability of the SARN; although this can perhaps be expected given the raters were highly experienced in rating and dealing with those who have committed sexual offending. The second study gave acceptable to good levels of support for inter rater reliability.
•Tully, Browne and Craig (2015) tested another version of the tool commonly used by the National Offender Management Service in England and Wales, focusing on ‘treatment needs assessment’ based on the risk factors applicable to each individual assessed (SARN TNA). This tool was applied to 496 adult males over two and four year follow up periods. ROC was found not to be significant at 2 years (AUC .59 in a sample of 304) or 4 years (AUC .57 in a sample of 161). It is, therefore, recommended that the SARN TNA is not relied upon as a predictor of sexual reoffending.
RATED page updated: August 2019 © Risk Management Authority 2019
•The instrument itself is only normed on adult males with a previous offence history and is deemed inappropriate for the use with females and juveniles.
General Notes
•Webster et al. (2006) circumscribed a number of provisions for use of the SARN: it should be applied by experienced psychologists; its use should be carefully monitored and evaluated; users should have demonstrated reasonable inter rater reliability before using the tool without supervision.
3)social stability and supports which consists of employment, residence, social influences (McGrath, Cumming and Lasher, 2013).
Author / Publisher McGrath, Cumming and Lasher
Category Sexual Offending (Awaiting Validation)
•The SOTIPS is a dynamic measure to be used with adult males who have sexually offended. It is designed to augment the findings from static risk tools by allowing for the identification of changeable risk factors that could be targets for treatment and supervision interventions (Lasher and McGrath, 2017; Miner et al., 2018).
•Improvements have been made from the previous version published in 2012. The number of items were reduced from 22 to 16 and a few items were edited to make them clearer. Sample interview questions and several case examples with accompanying scoring explanations are now provided for each item to assist assessors with their scoring. A definition of ‘qualifying sexual offence’ was also added to the manual. The ‘qualifying sex offences’ are divided into Category A and Category B offences. Category A offences are illegal sexual behaviours committed against an identifiable child or a non consenting adult victim. This definition also extends to online solicitation and non contact offences like exhibitionism and voyeurism. Category B offences are convictions for illegal sexual offences where there is no identifiable victim (possessing child pornography) or both parties were consenting (statutory rape if there is an age difference of less than 3 years, consenting sex with another adult in a public place, soliciting a prostitution) (McGrath, Cumming and Lasher, 2013).
1)sexual deviance, composed of sexual offence responsibility, sexual behaviour, sexual attitudes, sexual interests and sexual risk management;
Name of Tool Sexual Offender Treatment Intervention and Progress Scale (SOTIPS)
•Sixteen risk factors relate to three broad domains:
Description
Year 2013
•Scoring is to be undertaken on intake and thereafter every six months. Items are scored on a 4 point scale ranging from minimal, no need for improvement, considerable need for improvement and very considered need for improvement. These are totalled into a range of between 0 48 points, divided into three risk groups: low (scores of 0 10); moderate (scores of 11 20); high (scores of 21 48). The intention of this is to provide an estimation of an individual’s overall level of dynamic risk and need for supervision and treatment (McGrath, Cumming and Lasher, 2013; Lasher and McGrath, 2017).
RATED page updated: August 2019 © Risk Management Authority 2019
•It can be used in combination with other risk assessment tools such as the VASOR 2 or Static 99R (McGrath, Cumming and Lasher, 2013).
2)criminalityItems; factors, containing criminal and rule breaking behaviour, attitudes, stage of change, cooperation with treatment, cooperation with community supervision, emotion management, problem solving, impulsivity;
•Evaluators should consider information from multiple sources when scoring an individual: interviews, information from relevant sources (treatment providers, probation officers, etc.), behavioural observations and psychological tests (McGrath, Cumming and Lasher, 2013).
•The SOTIPS development sample (n=759) consisted of adult males who had been convicted of one or more qualifying sexual offences and had committed at least one of these when they were eighteen years of age. SOTIPS scored a good level of inter rater reliability for single measures (ICC=.77) and average measures (ICC=.87). There was also acceptable inter rater reliability for each of the three factors: sexual deviance, ICCs of .68 and .81 for single and average respectively; criminality, ICC of .76 for single and .86 for average; social stability and supports of .69 and .82 for single and average measures. Moderate to good predictive accuracy was shown for sexual, violent and any recidivism at 1, 7 and 12 months after participants began treatment (AUCs ranged from .60 to .85). Combining the SOTIPS scores with those from the Static 99R yielded a better degree of predictive accuracy than either instrument alone (AUCs ranged from .67 to .89). (McGrath, Lasher and Cumming, 2012).
•An evaluation of SOTIPS implementation at sites in Maricopa County and New York City was carried out by Miner and colleagues (2018). The findings from focus groups held with users of the instrument were that the SOTIPS was easy to understand and its focus on sexual offending made it useful and effective for their work.
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
Assessor Qualifications
•The SOTIPS focuses on three factors relating to sexual recidivism: 1) sexual deviancy, including sexual interests, offending, attitudes and motivation to change behaviour; 2) criminality, consisting of general antisociality, impulsiveness and oppositional reactions to rules; 3) social stability and supports looking at dysfunctional coping and the development of social support mechanisms (McGrath, Lasher and Cumming, 2012).
•The third SOTIPS assessment showed greater ability predict to sexual reoffending than the initial assessment. The AUCs were .71 and .72 for the third assessment of single and average scores respectively; in contrast to the AUC of .63 for the single score and the AUC of .64 for the average score for the initial assessment (Miner et at., 2018).
Tool Development
The SOTIPS was designed to be used by clinicians, case workers and probation/parole officers. Users of the SOTIPS should have a basic understanding of risk principles and risk factors relating to sexual offending. Users should carefully read the manual and undertake the training, which includes scoring practice cases. Training also includes how to use the SOTIPS with a static risk measure like the Static 99R or VASOR 2 (McGrath, Cumming and Lasher, 2013).
•Inter rater reliability was tested for the single and average scores on the SOTIPS. In the site of Maricopa County, ICCs were .653 and .790 for single and average scores respectively. In New York City, this was tracked over time with higher ICCs being generated for scoring done within one month (ICC=.821 for single; ICC=.902 for average) compared to within a time period of two months (ICC=.784 for single; ICC=.879 for average). These findings show that the SOTIPS is sensitive to change over time (Miner et al., 2018).
•A follow up study by Lasher and McGrath (2017) using a selection of the original sample (n=563) found that those who did not go on to reoffend demonstrated a greater degree of change during their first year in treatment than those who did commit further sexual or violent offences.
18+
•A study by Miner and colleagues (2018) found that the recency of the SOTIPS assessment was associated with accuracy. This means that it is acceptable for supervision levels to be adjusted based on the most recent risk level.
RATED page updated: August 2019 © Risk Management Authority 2019
•Miner and colleagues (2018) found that the SOTIPS items appear to be inclusive of those relating to sexual offending and general criminality. Based on this, they maintained that the SOTIPS is potentially useful for measuring the dynamic risk factors that can predict reoffending, as well as for guiding decision making relating to interventions, supervision and dispositions.
•Training workshops use case studies to demonstrate how to score the SOTIPS and also how to use it alongside other tools like the VASOR 2 and Static 99R. A ‘Train the Trainer’ workshop is available for up to 12 trainees who have already completed SOTIPS training and undertaken at least 10 assessments on individuals who have committed sexual offences. Following this workshop, trainees will be authorised to train staff at their organisation (McGrath, Cumming and Lasher, 2013).
•Risk categories vary depending on whether the SOTIPS is used alone or in conjunction with another tool. If the SOTIPS is used independently, the categories are low, moderate and high. If it is used in combination with a static risk instrument like the VASOR 2 or Static 99R, the risk categories would be mixed: for instance, if an individual scored as low on the VASOR 2 but as high on the SOTIPS this would give an overall rating of moderate low (McGrath, Cumming and Lasher, 2013).
•Lasher and colleagues (2015) compared therapist and client assessment scores in the middle of a prison based treatment programme designed for adult males who had committed sexual offences. Although findings showed there were significant differences in SOTIPS scores between therapists and clients, the correlations were substantial for criminality (ICC=.71), moderate for social stability and supports factor (ICC=0.59) and fair for sexual deviance (ICC=0.23). The authors maintained that the SOTIPS provides a useful framework to allow therapists to engage their clients collectively to identify strengths, treatment needs and potential treatment programmes.
•The manual advises that when the SOTIPS is used in residential settings, a few of the items are scored to reflect the individual’s level of functioning for the six months prior to his placement in prison or other secure residential settings (McGrath, Cumming and Lasher, 2013).
General Notes
•Since the SOTIPS does not cover all of the factors linked to sexually abusive behaviour, it is recommended by the developers that other relevant tools and professional judgment should be used in the supervision and treatment planning process (McGrath, Cumming and Lasher, 2013).
•Implementation of the SOTIPS into an organisation needs to take into consideration the workload, workflow and decision making processes of those undertaking assessments and interpreting the results (Miner et al., 2018).
•The manual is available free of charge here: http://www.robertmcgrath.us/index.php/risk instruments/sotips/
•The development study found the VASOR 2 predicted sexual recidivism with moderate accuracy (AUC .74) within a group of those who had committed sexual offences (n=1581). It also demonstrated good inter rater reliability with an ICC of .88 (McGrath, Hoke and Lasher, 2013).
•The VASOR 2 has fewer items and simpler scoring instructions than the original VASOR. The items on the original instrument about force used during index sex offence and amenability to treatment were removed in the revision (McGrath et al., 2014).
This tool is designed to be scored early by clinicians, correctional caseworkers and probation and parole officers. Users should have an understanding of the risk factors relating to sexual offending recidivism and risk assessment.
18+ Assessor Qualifications
Category Sexual Offending (Awaiting Validation)
•The VASOR 2 is a 14 item static risk tool designed to assess risk amongst adult males who have been convicted of at least one sexual offence.
•It is designed to be scored by clinicians, correctional caseworkers and probation/parole officers.
Age Appropriateness
Year 2013
Users must read the manual and complete training including practice cases to optimise scoring accuracy and reliability.
Name of Tool Vermont Assessment of Sex Offender Risk 2 (VASOR 2)
•It is two fold in nature: an 11 item reoffence risk scale consisting of static risk factors including alcohol and drug use, residence and employment stability and treatment amenability; a 3 item Severity Items checklist looking at the intrusive nature, the harm caused to the victim and the vulnerability of the victim all for index sexual offence. An individual’s score should be updated if they commit a new sexual offence (McGrath, Hoke and Lasher, 2013).
Author / Publisher McGrath, Hoke and Lasher
RATED page updated: August 2019 © Risk Management Authority 2019
Description
Tool Development
•The qualifying offences for this instrument are ‘Category A’ ones involving illegal sexual behaviour perpetrated against a child or adult victim. It may be used in cases of ‘Category B’ offences, which are illegal actions but with no identifiable victims or consenting parties (e.g. statutory rape), but only if the individual already has a conviction for a ‘Category A’ offence (McGrath, Hoke and Lasher, 2013).
RATED page updated: August 2019 © Risk Management Authority 2019
•The VASOR 2 has only limited value for making decisions about allowing an individual who has committed sexual offences to reside with children. The authors recommend that professional judgment and other tools should be used in the decision making process (McGrath, Hoke and Lasher, 2013).
•The VASOR 2 was tested for its predictive accuracy on four meta analytic datasets from Canada and Vermont (n=1581). It was found to predict sexual recidivism amongst different types of sexual offending: child abusers (n=1067, AUC=.74), rapists (n=395, AUC=.77) and non contact offences (n=87, AUC=.69). Caution is urged, however, when using the scale with those who have committed non contact sexual offences, due to the low sample size and statistical power (McGrath et al., 2014).
•Good inter rater reliability was shown for the VASOR 2 when tested using two independent ratings of thirty cases (McGrath et al., 2014).
General Notes
Beech, A. R., Fisher, D. D. and Thornton, D. (2003) ‘Risk assessment of sex offenders.’ Professional Psychology: Research and Practice 34(4), 339 352. Access Here.
Craig, L. A., Browne, K. D., Stringer, I. and Hogue, T. E. (2008) ‘Sexual reconviction rates in the United Kingdom and actuarial risk estimates.’ Child Abuse & Neglect 32(1), 121 138. Access Grubin,Here
© Risk Management Authority 2019
Helmus, L., Hanson, R. K., Babchishin, K. M. and Thornton, D. (2015) ‘Sex offender risk assessment with the risk matrix 2000: validations and guidelines for combining with the STABLE 2007.’ Journal of Sexual Aggression 21(2), 136 157. Access Here
Bengtson, S. (2008) ‘Is newer better? A cross validation of the Static 2002 and the Risk Matrix 2000 in a Danish sample of sexual offenders.’ Psychology, Crime & Law 14(2), 85 106. Access Here
Langton, C. M., Hogue, T. E., Daffern, M., Mannion, A. and Howells, K. (2009) ‘Prediction of institutional aggression among personality disordered forensic patients using actuarial and structured clinical risk assessment tools: prospective evaluation of the HCR 20, VRS, Static 99, and Risk Matrix 2000.’ Psychology, Crime & Law 15(7), 635 659. Access Here
VALIDATED TOOLS
RM2000
Craig, L. A., Browne, K. D. and Stringer, l. (2004) ‘Comparing Sex Offender Risk Assessment Measures on a UK Sample.’ International Journal of Offender Therapy and Comparative Criminology 48(1), 7 27. Access Here.
Hanson, R. K. and Thornton, D. (2000) ‘Improving Risk Assessments for Sex Offenders: A Comparison of Three Actuarial Scales.’ Law and Human Behavior 24(1), 119 136. Access Here.
RATED page updated: August 2019
D. (1998) Sexual Offending Against Children: Understanding the Risk. London: Home Office Research Development and Statistics Directorate. Access Here
Kingston, D. A., Yates, P. M., Firestone, P., Babchishin, K. and Bradford, J. M. (2008) ‘Long Term Predictive Validity of the Risk Matrix 2000: A Comparison With the Static 99 and the Sex Offender Risk Appraisal Guide.’ Sexual Abuse 20(4), 466 484. Access Here
Barnett, G. D., Wakeling, H. C. and Howard, P. D. (2010) ‘An Examination of the Predictive Validity of the Risk Matrix 2000 in England and Wales.’ Sexual Abuse 22(4), 443 470. Access Here.
Howard, P. and Wakeling, H. (2019) ‘Comparing two predictors of sexual recidivism: the Risk Matrix 2000 and the OASys Sexual Reoffending Predictor [Ministry of Justice Analytical Series]. HMPPS: Unpublished report. [Not accessible]
Grubin, D. (2011) ‘A Large Scale Evaluation of Risk Matrix 2000 in Scotland.’ Sexual Abuse, 23(4), 419 433. Access Here.
Knight, R. A. and Thornton, D. (2007) Evaluating and Improving Risk Assessment Schemes for Sexual Recidivism: A Long Term Follow Up of Convicted Sexual Offenders. Washington, D. C.: United States Department of Justice. Access Here.
Wakeling, H. C., Howard, P. and Barnett, G. (2011) ‘Comparing the Validity of the RM2000 Scales and OGRS3 for predicting recidivism by internet sexual offenders.’ Sexual Abuse: A Journal of Research and Treatment 23(1), 146 168. Access Here
© Risk
Osborn, J., Elliott, I. A., Middleton, D. and Beech, A. R. (2010) ‘The use of actuarial risk assessment measures with UK internet child pornography offenders. Journal of Aggression, Conflict and Peace Research 2(3), 16 24. Access Here.
Thornton, D. and Helmus, M. (2015) ‘Adjusting Risk Matrix Categories for Older Age and Time Offense Free.’ ATSA Conference, 16th October. Montreal, Quebec. [Not accessible]
Pryboda, J., Tully, R. J. and Browne, K. D. (2015) ‘Is the Risk Matrix 2000 applicable to intellectually disabled sex offenders?’ Aggression and Violent Behavior 25(A), 184 190. Access Here.
RATED page updated: August 2019 Management Authority 2019
Nicholls, C. M. and Webster, S. (2014) ‘Sex offender management and dynamic risk: pilot evaluation of the Active Risk Management System (ARMS).’ Ministry of Justice Analytical Series. Access Here
Smid, W. J., Kamphuis, J. H., Weaver, E. C. and Van Beek, D. J. (2014) ‘A comparison of the predictive properties of nine sex offender risk assessment instruments.’ Psychological Assessment 26(3), 691 703. Access Here
Looman, J. and Abracen, J. (2010) ‘Comparison of Measures of Risk for Recidivism in Sexual Offenders.’ Journal of Interpersonal Violence 25(5), 791 807. Access Here.
Tully, R. J. and Browne, K. B. (2015) ‘Appraising the Risk Matrix 2000 Static Sex Offender Risk Assessment Tool.’ International Journal of Offender Therapy and Comparative Criminology 59(2), 21 224. Access Here.
Lindsay, W. R., Hogue, T. E., Taylor, J. L., Steptoe, L., Mooney, P., O’Brien, G., Johnston, S., Smith, A. H. W. (2008) Risk Assessment in Offenders With Intellectual Disability: A Comparison Across Three Levels of Security.’ International Journal of Offender Therapy and Comparative Criminology, 52(1), 90 111. Access Here.
Thornton, D. (2010) Scoring guide for Risk Matrix 2000.10/SVC. London, England: Ministry of Justice. [Not accessible]
Middleton, D. (2008) ‘From research to practice: the development of the internet sex offender treatment programme (i SOTP).’ Irish Probation Journal 5, 49 64. Access Here.
Lehmann, R. J. B., Thornton, D., Helmus, L. M. and Hanson, R. K. (2016) ‘Developing Nonarbitrary Metrics for Risk Communication: Norms for the Risk Matrix 2000.’ Criminal Justice Behavior 43(12), 1661 1687. Access Here
Wakeling, H. C., Mann, R. E. and Milner, R. J. (2011) ‘Interrater Reliability of Risk Matrix 2000/s.’ International Journal of Offender Therapy and Comparative Criminology 55(8), 1324 1337. Access Here
Parent, G., Guay, J. P. and Knight, R. A. (2011) ‘An assessment of long term risk of recidivism by adult sex offenders: one size doesn’t fit all.’ Criminal Justice and Behavior 38(2), 188 209. Access Here
RRASOR Beech, A. R., Fisher, D. D. and Thornton, D. (2003) ‘Risk assessment of sex offenders.’ Professional Psychology: Research and Practice 34(4), 339 352. Access Here.
Långström, N. (2004) ‘Accuracy of Actuarial Procedures for Assessment of Sexual Offender Recidivism Risk May Vary Across Ethnicity.’ Sexual Abuse: A Journal of Research and Treatment 16(2), 107 120. Access Here
RATED page updated: August 2019 Risk Management Authority 2019
Langton, C. M., Barbaree, H. E., Seto, M. C., Peacock, E. J., Harkins, L. and Hansen, K. T. (2007) ‘Actuarial Assessment of Risk for Reoffense Among Adult Sex Offenders: Evaluating the Predictive Accuracy of the Static 2002 and Five Other Instruments.’ Criminal Justice and Behavior 34(1), 37 59. Access Here.
Webb, L., Craissati, J. and Keen, S. (2007) ‘Characteristics of Internet child pornography offenders: A comparison with child molesters.’ Sexual Abuse: A Journal of Research and Treatment 19(4), 449 465. Access Here
Blacker, J., Beech, A. R., Wilcox, D. T. and Boer, D. P. (2011) ‘The assessment of dynamic risk and recidivism in a sample of special needs sexual offenders.’ Psychology, Crime & Law 17(1), 75 92. Access Here
Sjöstedt, G. and Långström, N. (2002) ‘Assessment of risk for criminal recidivism among rapists: A Comparison of four different measures.’ Psychology, Crime & Law 8(1), 25 40. Access Here
©
Rettenberger, M., Matthes, A., Boer, D. P. and Eher, R. (2010) ‘Prospective Actuarial Risk Assessment: A Comparison of Five Risk Assessment Instruments in Different Sexual Offender Subtypes.’ International Journal of Offender Therapy and Comparative Criminology, 54(2), 169 186. Access Here
Parent, G., Guay, J. P. and Knight, R. A. (2011) ‘An assessment of long term risk of recidivism by adult sex offenders: one size doesn’t fit all.’ Criminal Justice and Behavior 38(2), 188 209. Access Here.
Knight, R. A. and Thornton, D. (2007) Evaluating and Improving Risk Assessment Schemes for Sexual Recidivism: A Long Term Follow Up of Convicted Sexual Offenders. Washington, D. C.: United States Department of Justice. Access Here.
Hanson, R. K. (1997) The development of a brief actuarial risk scale for sexual offense recidivism. Ottawa, Ontario: Department of the Solicitor General of Canada. Access Here.
Hanson, R. K. and Morton Bourgon, K. E. (2009) ‘The accuracy of recidivism risk assessments for sexual offenders: A meta analysis of 118 prediction studies.’ Psychological Assessment 21(1), 1 21. Access Here
Craig, L. A., Browne, K. D. and Stringer, l. (2004) ‘Comparing Sex Offender Risk Assessment Measures on a UK Sample.’ International Journal of Offender Therapy and Comparative Criminology 48(1), 7 27. Access Here.
Looman, J. and Abracen, J. (2010) ‘Comparison of Measures of Risk for Recidivism in Sexual Offenders.’ Journal of Interpersonal Violence 25(5), 791 807. Access Here
RATED page updated: August 2019 Risk Management Authority 2019
Cortoni,SA07
Hart, S. D. and Boer, D. P. (2010) ‘Structured professional judgment guidelines for sexual violent risk assessment.’ In R. K. Otto and K. S. Douglas. Handbook of Violence Risk Assessment. New York, London: Routledge, 269 294. Access Here
Yates, H. M. (2005) A Review of Evidence Based Practice in the Assessment and Treatment of Sex Offenders. Mechanicsburg, PA: Office of Planning, Research, Statistics and Grants, Pennsylvania Department of Corrections. Access Here
R. Russell, K., Forrest, L., Milton, E., Savoie, V., Baron, E., Kirkland, J. and Stobie, S. (2016) Risk for Sexual Violence Protocol (RSVP): A real world study of the reliability, validity and utility of a structured professional judgement instrument in the assessment and management of sexual offenders in South East Scotland. Edinburgh: NHS Lothian Sex Offender Liaison Service. Access Here.
Eher, R., Rettenberger, M., Matthes, A. and Schilling, F. (2010) ‘Stable dynamic risk factors in child sexual abusers: The incremental predictive power of narcissistic personality traits beyond
Sutherland, A. A. (2010). Sexual violence risk assessment: An investigation into the interrater reliability of the RSVP in Scotland. Doctoral thesis. Glasgow, UK: Section of Psychological Medicine, University of Glasgow. Access Here
Darjee,RSVP
Sutherland, A.A., Johnstone, L., Davidson., K.M., Hart, S., Cooke, D.J., Kropp, P.R., Logan, C., Michie, C. and Stocks, R. (2012) ‘Sexual violence risk assessment: An investigation of the interrater reliability of professional judgements made using the risk of sexual violence protocol.’ International Journal of Forensic Mental Health 11(2), 119 133. Access Here.
F., Hanson, K. R. and Coache, M. È. (2010) ‘The Recidivism Rates of Female Sexual Offenders Are Low: A Meta Analysis.’ Sexual Abuse 22(4), 387 401. Access Here.
Hart, S. D., Kropp, P. R., Laws, D. R., Klaver, J., Logan, C. and Watt, K. A. (2003) The Risk for Sexual Violence Protocol (RSVP): Structured Professional Guidelines for Assessing Risk of Sexual Violence British Columbia, Canada Simon Fraser University: Mental Health, Law, and Policy Institute, Simon Fraser University. [Not accessible]
Watt, K. A., Hart, S., Wilson, C., Guy, L. and Douglas, K. S. (2006) ‘An evaluation of the Risk for Sexual Violence Protocol (RSVP) in high risk offenders: Inter rater reliability and concurrent validity.’ Paper presented at the Annual Meeting of the American Psychology Law Society, St Petersburg, Florida, USA. [Not accessible]
©
Watt, K. A. and Jackson, K. (2008) ‘Inter rater and structural reliabilities of the Risk for Sexual Violence Protocol.’ Paper presented at the Annual Meeting of the International Association of Forensic Mental Health Services, Vienna, Austria. [Not accessible]
Hart, S. D., Michie, C. and Cooke, D. J. (2007) ‘Precision of actuarial risk assessment instruments: Evaluating the ‘margins of error’ of group v. individual predictions of violence.’ British Journal of Psychiatry 190 (suppl. 49), s60 s65. Access Here
Jackson, K. J. (2016) Validation of the Risk for Sexual Violence Protocol in Adult Sexual Offenders. Doctoral thesis. British Columbia, Canada Simon Fraser University. Access Here.
Hanson, K. R., Helmus, L. M. and Harris, R., A. J. (2015) ‘Assessing the Risk and Needs of Supervised Sexual Offenders: A Prospective Study Using STABLE 2007, Static 99R, and Static 2002R.’ Criminal Justice and Behavior 42(12), 1205 1224. Access Here
Tamatea, Amran J. (2014) ‘Predictive validity of the Stable 2007: A New Zealand Study.’ Sexual Abuse in Australia and New Zealand 6(1), 57 72. Access Here
Fabian, J. (2011) ‘Assessing the Sex Offender With Asperger’s Disorder: A Forensic Psychological and Neuropsychological Perspective.’ Sex Offender Law Report 12(5), 65 80. [Not Fernandez,accessible]Y.(2008)
Hanson, R. K., Harris, A. J. R., Scott, T. and Helmus, L. (2007) Assessing the risk of sexual offenders on community supervision: The Dynamic Supervision Project. Ontario, Canada: Public Safety Canada. Access Here.
RATED page updated: August 2019
Eher, R, Mattes, A, Schilling, F, Haubner MacLean, T & Rettenberger, M (2011) ‘Dynamic Risk Assessment in Sexual Offenders Using STABLE 2000 and the STABLE 2007: An Investigation of predictive and Incremental Validity.’ Sexual Abuse: A Journal of Research and Treatment 20(10), 1 24. Access Here.
Thornton, D. (2002) ‘Constructing and Testing a Framework for Dynamic Risk Assessment.’ Sexual Abuse: A Journal of Research and Treatment 14 (2), 139 153. Access Here
the Static 99/Stable 2007 priority categories on sexual reoffense.’ Sex Offender Treatment 5(1). Access Here.
Fernandez, Y., Harris, A. J. R., Hanson, R. K. and Sparks, J. (2012) STABLE 2007 coding manual revised 2012. Ottawa, ON: Public Safety Canada. [Not accessible]
Eher, R., Olver, M. E., Heurix, I., Schilling, F. and Rettenberger, M. (2015) ‘Predicting reoffence in pedophilic child molesters by clinical diagnoses and risk assessment.’ Law and Human Behavior 39(6), 571 580. Access Here
Mann, R. E., Hanson, R. K. and Thornton, D. (2010) ‘Assessing risk for sexual recidivism: Some proposals on the nature of psychologically meaningful risk factors.’ Sexual Abuse: A Journal of Research and Treatment 22(2), 191 217. Access Here.
Smid, W. J., Kamphuis, J. H., Weaver, E. C. and Van Beek, D. J. (2014) ‘A comparison of the predictive properties of nine sex offender risk assessment instruments.’ Psychological Assessment 26(3), 691 703. Access Here.
McNaughton Nicholls, C., Callanan, M., Legard, R., Tomaszewski, W., Purdon, S. and Webster, S.(2010) ‘Examining implementation of the Stable and Acute dynamic risk assessment tool pilot in England and Wales.’ Research Series 4/10. London: Ministry of Justice. Access Here
© Risk Management Authority 2019
An examination of the inter rater reliability of the STATIC 99 and STABLE 2007. Poster presentation at the 27th Annual Research and Treatment conference of the Association for the Treatment of Sexual Abusers, Atlanta, GA, in October. [Not accessible]
Sowden, J. N. and Olver, M. E. (2017) ‘Use of the violent risk scale sexual offender version and the Stable 2007 to assess dynamic sexual violence risk in a sample of treated sexual offenders.’ Psychological Assessment 29(3), 293 303. Access Here
Rettenberger, M. and Eher, R. (2007) ‘Predicting reoffence in sexual offender subtypes: A prospective validation study of the German version of the Sexual Offender Risk Appraisal Guide (SORAG).’ Sexual Offender Treatment, 2(2), 1 12. Access Here
Rettenberger, M., M. E. Rice, G. T. Harris and R. Eher. (2017) ‘Actuarial risk assessment of sex offenders: the psychometric properties of the Sex Offender Risk Appraisal Guide (SORAG).’ Psychological Associaiton 29(6), 624 638. Access Here
Walker, M. and O’ Rourke M. (2013) ‘Probation Officers’ Experience of Using Risk Matrix 2000 and Stable and Acute 2007 when supervising sex offenders living in the community.’ Irish Probation Journal 10, 162 176. Access Here
Eher, R., Rettenberger, M., Schilling, F. and Pfäfflin, F. (2008) ‘Failure of Static 99 and SORAG to predict relevant reoffense categories in relevant sexual offender subtypes: A prospective study.’ Sexual Offender Treatment 3(1). Access Here
Ducrol, C. and Pham, T. (2006) ‘Evaluation of the SORAG and the Static 99 on Belgian Sex Offenders Committed to a Forensic Facility.’ Sexual Abuse 18(1), 15 26. Access Here.
RATED page updated: August 2019 Management
Authority 2019
Langton, C. M., Hogue, T. E., Daffern, M., Mannion, A. and Howells, K. (2009) ‘Prediction of institutional aggression among personality disordered forensic patients using actuarial and structured clinical risk assessment tools: prospective evaluation of the HCR 20, VRS, Static 99, and Risk Matrix 2000.’ Psychology, Crime & Law 15(7), 635 659. Access Here
Looman, J. (2006) ‘Comparison of Two Risk Assessment Instruments for Sexual Offenders.’ Sexual Abuse 18(2), 193 206. Access Here
Parent, G., Guay, J. P. and Knight, R. A. (2011) ‘An assessment of long term risk of recidivism by adult sex offenders: one size doesn’t fit all.’ Criminal Justice and Behavior 38(2), 188 209. Access Here
Rettenberger, M., Matthes, A., Boer, D. P. and Eher, R. (2010) ‘Prospective Actuarial Risk Assessment: A Comparison of Five Risk Assessment Instruments in Different Sexual Offender Subtypes.’ International Journal of Offender Therapy and Comparative Criminology, 54(2), 169 186. Access Here.
SORAG
© Risk
Pham, T. H. and Ducro, C. (2008) ‘Évaluation du risque de récidive en Belgique francophone: Données préliminaires d'analyse factorielle de la “Sex Offender Recidivism Appraisal Guide” (SORAG) et de la Statique 99 [Risk assessment in social defence: Preliminary factorial analysis of the Sex Offender Recidivism Appraisal Guide (SORAG) and the Static 99].’ Annales Médico Psychologiques 166(7), 575 579. Access Here.
Nunes, K. L., Firestone, P., Bradford, J. M., Greenberg, D. M. and Broom, I. (2002) ‘A comparison of modified versions of the Static 99 and the Sex Offender Risk Appraisal Guide.’ Sexual Abuse: A journal of Research and Treatment 14(2), 253 269. Access Here.
Rice, M. E. and G. T. Harris. (2016) ‘The Sex Offender Risk Appraisal Guide.’ In Phenix, A. and Hoberman, H. (eds.) Sexual Offending: Predisposing Antecedents, Assessments and Management. New York: Springer, 471 488. Access Here
Babchishin, K. M., Hanson, R. K. and Helmus, L. (2012) ‘Communicating risk for sex offenders: risk ratios for Static 2002R.’ Sexual Offender Treatment 7(2), 1 12. Access Here.
Hanson, K. R., Harris, R., A. J., Helmus, L. and Thornton, D. (2014) ‘High Risk Sex Offenders May Not Be High Risk Forever.’ Journal of Interpersonal Violence 29(15), 2792 2813. Access Here
Brouillette Alarie, S., Proulx, J. and Hanson, R. K. (2017)’Three central dimensions of sexual recidivism risk: understanding the latent constructs of the Static 99R and Static 2002R.’ Sexual Abuse: A Journal of Research and Treatment 30(6), 676 704. Access Here
Hanson, R. K., Babchishin, K. M., Helmus, L. M., Thornton, D. and Phenix, A. (2017) ‘Communicating the results of criterion referenced prediction measures: Risk categories for the Static 99R and Static 2002R sexual offender risk assessment tools.’ Psychological Assessment 29(5), 582 597. Access Here
RATED page updated: August 2019
Jung, S., Ennis, L., Hermann, C. A., Pham, A. T., Choy, A. L., Corabian, G. and Hock, T. (2017) ‘An evaluation of the reliability, construct validity and actor structure of the Static 2002R.’ International Journal of Offender Therapy and Comparative Criminology 61(4), 464 487. Access Here
Walters, G. D., Knight, R. A. and Thornton, D. (2009) ‘The latent structure of sexual violence risk: A taxometric analysis of widely used sex offender actuarial risk measures.’ Criminal Justice and Behavior 36(3), 290 306. Access Here
Rossegger, A.; Gerth, J.; Singh, J. P. and Endrass, J. (2013) ‘Examining the predictive validity of the SORAG in Switzerland.’ Sex Offender Treatment 8(2), 1 12. Access Here.
Yates, H. M. (2005) A Review of Evidence Based Practice in the Assessment and Treatment of Sex Offenders. Mechanicsburg, PA: Office of Planning, Research, Statistics and Grants, Pennsylvania Department of Corrections. Access Here.
Quinsey, L., Harris, G. T., Rice, M. E. and Cormier, C. A. (2006) Violent Offenders: Appraising and Managing the Risk. Washington, D. C.: American Psychological Association. [Not accessible]
M., Hanson, R. K. and Helmus, L. (2011) The RRASOR, Static 99R, and Static 2002R All Add Incrementally to the Prediction of Recidivism among Sex Offenders. Corrections research user report no. 2011 02. Ottawa, Canada: Public Safety Canada. Access Here.
Hanson, R. K., Harris, A. J., Letourneau, E., Helmus, L. M. and Thornton, D. (2018) ‘Reductions in risk based on time offense free in the community: Once a sexual offender, not always a sexual offender.’ Psychology, Public Policy and Law 24(1), 48 63. Access Here.
Helmus, L. (2009) Re norming Static 99 recidivism estimates: Exploring base rate variability across sex offender samples. Unpublished M.A. thesis. Ontario, Canada: Carleton University. Access Helmus,HereL.,Thornton, D., Hanson, R. K., & Babchishin, K. M. (2012). Improving the predictive accuracy of Static 99 and Static 2002 with older sex offenders: Revised age weights. Sexual Abuse: A Journal of Research and Treatment 24(1), 64 101. Access Here.
Static Babchishin,2002RK.
© Risk Management Authority 2019
Kanters, T., Hornsveld, R. H. J., Nunes, K. L., Zwets, A. J., Muris, P. and Van Marie, H. J. C. (2017) ‘The Sexual Violence Risk 20: factor structure and psychometric properties.’ The Journal of Forensic Psychiatry & Psychology 28(3), 368 387. Access Here
© Management Authority 2019
SVR Blacker,20
RATED page updated: August 2019
Rohrer, D. M. (2019) Replication of a Three Factor Solution: Exploring the Underlying Constructs of the Static 99R and Static 2002R. Master of Arts thesis. Brandeis University: Waltham, Massachusetts. Access Here.
Parent, G., Guay, J. P. and Knight, R. A. (2011) ‘An assessment of long term risk of recidivism by adult sex offenders: one size doesn’t fit all.’ Criminal Justice and Behavior 38(2), 188 209. Access Phenix,Here.A.,Doren,
D., Helmus, L., Hanson, R. K. and Thornton, D. (2009) Coding rules for Static 2002. Ottawa, ON: Public Safety Canada. Access Here.
Craig, L. A., Beech, A. and Browne, K. D. (2006a) ‘Cross validation of the Risk Matrix 2000 Sexual and Violent scales.’ Journal of Interpersonal Violence, 21(5), 612 633. Access Here.
Hill, A., Habermann, N., Klusmann, D., Berner, W. and Briken, P. (2008) ‘Criminal Recidivism in Sexual Homicide Perpetrators.’ International Journal of Offender Therapy and Comparative Criminology 52(1), 5 20. Access Here
Risk
De Vogel, V., de Ruiter, C., van Beek, D. and Mead, G. (2004) ‘Predictive validity of the SVR 20 and Static 99 in a Dutch sample of treated sex offenders.’ Law and Human Behavior 28(3), 235 251. Access Here.
Lee, S. C. (2019) Cross Cultural Validity of Actuarial Risk Assessment Instruments for Individuals in North America with a History of Sexual Offending: Static 99R and Static 2002R. Doctoral thesis. Carleton University: Ottawa, Ontario. Access Here
Craig, L. A., Beech, A. and Browne, K. D. (2006b). ‘Evaluating the Predictive Accuracy of Sex Offender Risk Assessment Measures on UK Samples: A Cross Validation of the Risk Matrix 2000 Scales.’ Sexual Offender Treatment 1(1). Access Here
Reeves, S. G., Ogloff, J. R. P. and Simmons, M. (2017) ‘The predictive validity of the Static 99, Static 99R and Static 2002/r: which one to use?’ Sexual Abuse: A Journal of Research and Treatment 30(8), 887 907. Access Here
Hart, S. D. and Boer, D. P. (2010) ‘Structured professional judgment guidelines for sexual violent risk assessment.’ In R. K. Otto and K. S. Douglas. Handbook of Violence Risk Assessment. New York, London: Routledge, 269 294. Access Here.
Dietiker, J., Dittmann, V. and Graf, M. (2007) ‘Risk assessment of sex offenders in a German speaking sample. applicability of PCL SV, HCR 20+3, and SVR 20.’ Nervenarzt 78(1), 53 61. Access Here
J., Beech, A. R., Wilcox, D. T. and Boer, D. P. (2011) ‘The assessment of dynamic risk and recidivism in a sample of special needs sexual offenders.’ Psychology, Crime & Law 17(1), 75 92. Access Here
T., Di Placido, C. and Wong, S. (2001) How dangerous are dangerous sex offenders? An estimation of recidivism and level of risk using a matched control group. Saskatoon, Saskatchewan, Canada: Regional Psychiatric Centre. [Not accessible]
Knight, R. A. and Thornton, D. (2007) Evaluating and Improving Risk Assessment Schemes for Sexual Recidivism: A Long Term Follow Up of Convicted Sexual Offenders. Washington, D. C.: United States Department of Justice. Access Here
Olver, M. E., Beggs Christofferson, S. M., Grace, R. C., and Wong, S. C. P. (2014) ‘Incorporating Change Information Into Sexual Offender Risk Assessments Using the Violence Risk Scale Sexual Offender Version.’ Sexual Abuse 26(5), 472 499. Access Here
RATED page updated: August 2019
Olver, M. E. (2004) ‘The development and validation of the Violence Risk Scale: Sexual Offender version (VRS:SO) and its relationship to psychopathy and treatment attrition.’ Dissertation Abstracts International: Section B: The Sciences and Engineering 64(9 B), 4628. Access Here
Rettenberger, M., Boer, D. P. and Eher, R. (2011) ‘The Predictive Accuracy of Risk Factors in the Sexual Violence Risk 20 (Svr 20).’ Criminal Justice and Behavior 38(10), 1009 1027. Access
VRS:SO
Beggs, S. M. and Grace, R. C. (2010) ‘Assessment of Dynamic Risk Factors: An Independent Validation Study of the Violence Risk Scale: Sexual Offender Version.’ Sexual Abuse 22(2), 234 251. Access Here
G. and Långström, N. (2002) ‘Assessment of risk for criminal recidivism among rapists: A Comparison of four different measures.’ Psychology, Crime & Law 8(1), 25 40. Access
Ramírez, M. P., Illescas, S. R., García, M. M., Forero, C. G. and Pueyo, A. A. (2008) ‘Predicción de riesgo de reincidencia en agresores sexuales [Recidivism risk assessment in sex offenders].’ Psicothema 20(2), 205 210. Access Here
© Risk Management Authority 2019
Sjöstedt,Here
Rettenberger, M., Matthes, A., Boer, D. P. and Eher, R. (2010) ‘Prospective Actuarial Risk Assessment: A Comparison of Five Risk Assessment Instruments in Different Sexual Offender Subtypes.’ International Journal of Offender Therapy and Comparative Criminology, 54(2), 169 186. Access Here.
Witte,Here.
Maltais, N. and Sribney, C. (2018) Assessing the Convergent Validity of the VRS:SO and the CPORT with a Community Sample. Poster presentation at the Annual Conference of the Association for the Treatment of Sexual Abusers, Vancouver, B.C., 17 19 October. Access Here.
Olver, M. E., Wong, S. C. P., Nicholaichuk, T. and Gordon, A. (2007) ‘The validity and reliability of the Violence Risk Scale Sexual Offender version: Assessing sex offender risk and evaluating therapeutic change.’ Psychological Assessment 19(3), 318 329. Access Here.
Parent, G., Guay, J. P. and Knight, R. A. (2011) ‘An assessment of long term risk of recidivism by adult sex offenders: one size doesn’t fit all.’ Criminal Justice and Behavior 38(2), 188 209. Access Here.
M., Hanson, R. K. and Blais, J. (2013) User Guide for the Brief Assessment for Recidivism Risk 2002R (BARR 2002R). Based on a paper presented by K. Babchishin at the 32nd Annual Research and Treatment Conference of the Association for the Treatment of Sexual Abusers, October 30th, Chicago, Illinois. Access Here
TOOLS AWAITING VALIDATION BARR Babchishin,2002RK.
J. N. (2013) Examining the Relationship of Risk, Treating Readiness and Therapeutic Change to Recidivism in a Sample of Treated Sex Offenders. Doctoral thesis. Saskatoon, Saskatchewan: University of Saskatchewan. Access Here
Sowden, J. N. and Olver, M. E. (2017) ‘Use of the violent risk scale sexual offender version and the Stable 2007 to assess dynamic sexual violence risk in a sample of treated sexual offenders.’ Psychological Assessment 29(3), 293 303. Access Here.
Sowden,Here.
Eke,CPORTA.
W., Maaike Helmus, L. and Seto, M. C. (2018) ‘A Validation Study of the Child Pornography Offender Risk Tool (CPORT).’ Sexual Abuse Access Here
Jung, S. and Wielinga, F. (2019) ‘Simplifying the estimation of violence risk by police among individuals charged for sexual assault.’ Journal of Threat Assessment and Management 6(1), 38 50. Access Here.
Olver, M. E., Neumann, C. S., Kingston, D. A., Nicholaichuk, T. P. and Wong, S. C. P. (2016) ‘Construct validity of the violence risk scale sexual offender version instrument in a multisite sample of treated sexual offenders.’ Assessment 25(1), 40 55. Access Here.
Olver, M. E., Neumann, C. S., Kingston, D. A., Nicholaichuk, T. P. and Wong, S. C. P. (2018b) ‘Construct validity of the Violence Risk Scale Sexual Offender Version instrument in a multisite sample of treated sexual offenders.’ Assessment 25(1), 40 55. Access Here
Olver, M. E., Mundt, J. C., Thornton, D., Beggs Christofferson, S. M., Kingston, D. A., Sowden, J. N., Nicholaichuck, T. P., Goprdon, A., and Wong, S. C. P. (2018a) ‘Using the Violence Risk Scale Sexual Offence Version in Sexual Violence Risk Assessments: updated risk categories and recidivism estimates from a multisite sample of treated sexual offenders.’ Psychological Assessment 30(7), 941 955. Access Here.
Olver, M. E., Nicholaichuk, T. P., Kingston, D. A. and Wong, S. C. P. (2014) ‘A multisite examination of sexual violence risk and therapeutic change.’ Journal of Consulting and Clinical Psychology 82(2), 312 324. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Jung, S., Wielinga, F. and Ennis, L. (2018) ‘Can we keep it simple? Using the Barr 2002R with a community based sex offender sample.’ Journal of Sexual Aggression 24(1), 25 36. Access Here
Babchishin, K. M., Hanson, R. K. and Blais, J. (2016) ‘Less is more: using Static 2002R subscales to predict violent and general recidivism among sexual offenders.’ Sexual Abuse: A Journal of Research and Treatment 28(3), 187 217. Access Here.
Prochaska, J. O., DiClemente, C. C. and Norcross, J. C. (1992) ‘In search of how people change: Applications to the addictive behaviors.’ American Psychologist 47(9), 1102 1114. Access
Long, M. L., Alison, L. A. and McManus, M. A. (2013) ‘Child Pornography and Likelihood of Contact Abuse: A Comparison Between Contact Child Sexual Offenders and Noncontact Offenders.’ Sexual Abuse 25(4), 370 395. Access Here.
Gunnarsdóttir, H. Ó. (2019) Risk Assessment of convicted Child Pornography Offenders in Iceland 2000 2017. Master of Clinical Psychology Thesis. Iceland: Skemann University. Access Maltais,Here.
Beech,SARN
RATED page updated: August 2019
Hocken, K., Winder, B. and Grayson, A. (2013) ‘Putting responsivity into risk assessment: the use of the Structured Assessment of Risk and Need (SARN) with sexual offenders who have an intellectual disability.’ Journal of Intellectual Disabilities and Offending 4(3/4), 77 89. Access Here
Long, M., Alison, L., Tejeiro, R., Hendricks, E. and Giles, S. (2016) ‘KIRAT: Law Enforcement’s Prioritization Tool for Investigating Indecent Image Offenders.’ Psychology, Public Policy and Law 22(1), 12 21. Access Here
N. and Sribney, C. (2018) Assessing the Convergent Validity of the VRS:SO and the CPORT with a Community Sample. Poster presentation at the Annual Conference of the Association for the Treatment of Sexual Abusers, Vancouver, B.C., 17 19 October. Access Here
Seto, M. C. and Eke, A. W. (2017).’Correlates of Admitted Sexual Interest in Children Among Individuals Convicted of Child Pornography Offenses.’ Law and Human Behavior 41(3), 305 313. Access Here
McManus, M.A., Long, M. L. and Alison, L. (2011) ‘Child pornography offenders: towards an evidenced based approach to prioritizing the investigation of indecent image offences.’ In Alison, L. and Rainbow, L. Professionalizing Offender Profiling: Forensic and Investigative Psychology in Practice. London: Routledge, 178 188. Access Here
© Risk Management Authority 2019
Pilan, A. J. M. (2016) The predictive validity of general and offence specific risk assessment tools for child pornography offenders’ reoffending. Degree of Master of Art thesis. Saskatoon, Canada: University of Saskatchewan. Access Here.
Eke, A. W. and Seto, M. C. (2016) Scoring guide for the child pornography offender risk tool (CPORT). Access Here
Seto, M. C. and Eke, A. W. (2015) ‘Predicting recidivism among adult male child pornography offenders: development of the child pornography offender risk tool (CPORT).’ Law and Human Behavior 39(4), 416 429. Access Here
Child Exploitation and Online Protection Centre. (2012). A picture of abuse: a thematic assessment of the risk of contact child sexual abuse posed by those who possess indecent images of children. London: CEOP. Access Here
KIRAT 2
A. R., Fisher, D. D. and Thornton, D. (2003) ‘Risk assessment of sex offenders.’ Professional Psychology: Research and Practice 34(4), 339 352. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Lasher,SOTIPS
Smid, W. J., Kamphuis, J. H., Weaver, E. C. and Van Beek, D. J. (2014) ‘A comparison of the predictive properties of nine sex offender risk assessment instruments.’ Psychological Assessment 26(3), 691 703. Access Here
Tully, R. J., Browne, K. D. and Craig, L. A. (2015) ‘An examination of the predictive validity of the structured assessment of the risk and needs treatment needs analysis (SARN TNA) in England and Wales.’ Criminal Justice & Behavior 42(5), 509 528. Access Here.
Lasher, M. P. and McGrath, R. J. (2017) ‘Desistance from sexual and other violent offending among child sexual abusers: observations using the Sex Offender Treatment Intervention and Progress Scale (SOTIPS).’ Criminal Justice and Behavior 44, 416 431. Access Here.
M., McGrath, R. J., Wilson, D., and Cumming, G. F. (2015) ‘Collaborative Treatment Planning Using the Sex Offender Treatment Intervention and Progress Scale (SOTIPS): Concordance of Therapist Evaluation and Client Self Evaluation.’ Internal Journal of Forensic Mental Health 14(1), 1 9. Access Here
Cumming, G. F. and Lasher, M. P. (2013) SOTIPS (Sex Offender Treatment Intervention and Progress Scale) Manual. Middlebury: VT. Access Here.
Webster, S. D., Mann, R. E., Carter, A. J., Long, J., Milner, R. J. O’Brien, M. D., Wakeling, H. C. and Ray, N. L. (2006) ‘Inter rater reliability of dynamic risk assessment with sexual offenders.’ Psychology, Crime and Law 12(4), 439 452. Access Here
McGrath, R. J., Lasher, M. P. and Cumming, G. F. (2012) ‘The Sex Offender Treatment Intervention and Progress Scale (SOTIPS): Psychometric properties and incremental predictive validity with Static 99R.’ Sexual Abuse: A Journal of Research and Treatment 24(5), 431 458. Access McGrath,Here.R.J.,
Miner, M. H., Robinson, B. E., Newstrom N. and Strobel Ayres, C. (2018) Evaluation of the Implementation of the Sex Offender Treatment Intervention and Progress Scale (SOTIPS). National Institute of Justice: Washington, D. C. Access Here
VASOR 2
McGrath, R. J., Hoke, S. E. and Lasher, M. P. (2013) VASOR 2: Vermont Assessment of Sex Offender Risk 2 Manual. Middlebury, VT. Access Here.
McGrath, R. J., Lasher, M. P., Cumming, G. F., Langton, C. M. and Hoke, S. E. (2014) ‘Development of Vermont Assessment of Sex Offender Risk 2 (VASOR 2) Reoffense Risk Scale.’ Sexual Abuse: A Journal of Research & Treatment 26(3), 271 290. Access Here
Thornton, D. (2002) ‘Constructing and Testing a Framework for Dynamic Risk Assessment.’ Sexual Abuse: A Journal of Research and Treatment 14 (2), 139 153. Access Here
18+
•The CAPP SRS uses information derived from file review, a detailed clinical interview (CAPP SRS Clinical Interview) and information obtained from an informant using the CAPP SRS Informant Report (Cooke and Logan, 2015; Cooke and Logan, 2018; Cooke et al., under review)
Year 2004 Description
CAPP SRS training is suitable for experienced practitioners in the fields of psychiatry and psychology who are already trained in the assessment of psychopathy and who use structured assessments of personality disorder in their work with clients or research participants in forensic hospital or correctional settings (Cooke et al., 2004).
•The CAPP SRS (originally named the CAPP IRS) is a measure of psychopathic disorder (PPD), based on the Comprehensive Assessment of Personality conceptual model of PPD (Cooke et al., 2012). It is an expert observer symptom rating scale suitable for use in clinical and forensic settings (Cooke and Logan, 2015; Cooke and Logan, 2018)
RATED page updated: August 2019 © Risk Management Authority 2019
Category Responsivity Issues (Validated)
•The CAPP SRS consists of 33 symptoms grouped across six domains. Seven point scale ratings are given for each symptom (Florez et al., 2018).
Assessor Qualifications
Strengths
Age Appropriateness
•The tool specifically focuses on the domain of personality pathology rather than mixing the domains of personality pathology with criminal or anti social behaviours; it is thus less tautological than other measures of PPD (Cooke and Sellbom, 2019; Skeem and Cooke, 2010).
Author / Publisher Cooke, Hart, Logan and Michie
Assessor should have experience and training in administering and interpreting assessments of personality disorder. In addition, assessor should have training in the application of the CAPP SRS
•The CAPP SRS provides a detailed and idiographic description of psychopathic traits that are known to be linked to violence risk. Symptoms of PPD are allocated to one of six domains of basic personality functioning i.e., self, attachment, behavioural, cognitive, dominance and emotional. The comprehensive nature of the measure provides the foundation for detailed and nuanced diagnostic and risk formulations of the individual case (Cooke and Logan, 2015; Cooke and Logan, 2018).
Name of Tool The Comprehensive Assessment of Psychopathic Personality Symptom Rating Sale (CAPP SRS)
Empirical Grounding
No empirical research at present.
•Prototypicality studies reveal striking consistency in which symptoms are regarded as most central or diagnostic of PPD (Cooke, 2018).
To ensure complete coverage of the construct domains, the map also included some symptoms that were controversial and not identified by SMEs or in the research literature. Symptoms that were virtually synonymous were grouped together to give a set of 33 symptoms (Cooke, 2018; Cooke, Hart and Michie, under review).
RATED page updated: August 2019 © Risk Management Authority 2019
b)International Research
•A Masters of Arts dissertation tested the inter rater reliability of the CAPP in a sample of 30 incarcerated youth. The total CAPP scores had an excellent overall IRR of 0.91. The domain IRR scores ranged from good to excellent (ICC=0.69 to 0.86). The rating of 0.69 is believed to be due to fair ICCs of .50 and below for some of the self domain symptoms: self centred, sense of entitlement, sense of invulnerability and unstable self concept. This suggests that items may be more difficult for raters to assess consistently (McCormick, 2004).
•De Page, Mercenier and Titeca (2018) tested the CAPP IRS (the former name for the CAPP SRS) in a sample of 72 male forensic patients with a primary diagnosis of schizophrenia spectrum disorders. The CAPP IRS showed good inter rater reliability.
•Using a Norwegian research version of the CAPP IRS (now the CAPP SRS) on eighty male inmates, Sandvik and colleagues (2012) found that the inter rater reliability
•Sea (2018) utilised a Korean translation of the CAPP SRS in a sample of correctional offenders in South Korea. The inter rater reliability scores were very high for almost all of the symptoms, ranging from .82 to .90. Concurrently validity was also demonstrated, with K CAPP SRS total scores correlating highly with total scores on the Korean translation of the PCL R.
•Concept maps are a means to explicitly lay out knowledge about a particular topic in a form that is simple and graphical. The concept map was developed based on reviews of the clinical and research literatures looking at existing diagnostic criteria and detailed clinical and research descriptions of psychopathic personality disorder (PPD). Detailed interviews were also undertaken with subject matter experts (SMEs): clinicians working closely with patients with PPD (Cooke, 2018).
Inter Rater Reliability
a)UK Research
b)International Research
•Using a sample of patients in a forensic unit in Denmark, Pedersen et al. (2010) found that IRR for the CAPP ranged from fair/good through to excellent. The total ICC was .56. The domains varied: attachment, ICC=.89; behavioural, ICC=.76; cognitive, ICC=.74; dominance, ICC=.92; emotional, ICC=.88; self, ICC=.87.
a)UK Research
•The CAPP SRS was applied in a sample of 204 Spanish prisoners and was a found to ‘robust and solid method’ to evaluate psychopathy in a correctional setting (Florez et al., 2018).
•Pedersen et al. (2010) the CAPP achieved moderate accuracy (AUC) in predicting violent (.70) and non violent (.71) recidivism in a 5 year follow up with forensic psychiatric patients, similar to the predictive accuracy observed for the PCL:SV.
•McCuish and colleagues (2019) tested the inter rater reliability of the CAPP IRS. IRR was excellent for the total scores (IICC=.0.91) and adequate to excellent for domain scores (0.69 0.86).
•Cooke and colleagues (under review) reported on a seven site study 315 adult male correctional offenders and secure hospital patients in Scotland and England and found that the CAPP SRS had good measurement precision and good external validity with respect to scores on an older test of PPD, the Hare Psychopathy Checklist Revised (Hare, 1991, 2003).
General Predictive Accuracy
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
ranged from good to excellent. The Total score yielded an ICC of .97. The domains gave the following results: attachment, IC=.89; behavioural, ICC=.76; cognitive, ICC=.74; dominance, ICC=.92; emotional, ICC=.88; self, ICC=.87.
•Inter rater reliability for the CAPP IRS was found to be excellent for total scores (ICC=0.91) and range from adequate to excellent for domain scores (ICC=0.69 0.86) in a study involving adolescents from the Incarcerated Serious Violent Youth Offender study (McCuish, Hanniball and Corrado, 2019).
Validation History
b)International Research
No empirical research at present.
RATED page updated: August 2019 © Risk Management Authority 2019
b)International Research
a)UK Research
Validation History
•A doctoral thesis by Kreis (2009) looked at the CAPP in sample of women offenders (n=20) using semi structured interviews and self report. The conclusion was reached that at a symptom level prototypical psychopathic women and men and are very similar; although important gender differences do exist, particularly in the expression of symptoms. The CAPP was found to capture psychopathy well across the female gender.
Validation History
•Kreis and Cooke (2012) applied the CAPP IRS to two case studies of female offenders. It showed promise for use with women, allowing for greater exploration of nuances in traits. The authors caution, however, that it is still under validation and there are no norms available for using the CAPP IRS with females.
Applicability: Females
a)UK Research
•Pauli and colleagues (2018) administered questionnaires to correctional officers in Sweden who rate male or female psychopathy to test whether the CAPP IRS symptoms were applicable across both genders. Most of the CAPP symptoms were rated as highly or moderately typical of both female and male psychopathy; although female participants in the study rated ‘Domineering’ as significantly more typical of psychopathy than the male officers did. Although the study downed that CAPP symptoms are relatively gender neutral, there were some differences in how psychopathy symptoms were described between the genders: psychopathic men were described as reckless, uncaring, self aggrandising, emotional expressive and garrulous; whereas women were described as more detached and lacking pleasure.
No empirical research at present.
Applicability: Ethnic Minorities
•The CAPP SRS was originally named the CAPP Institutional Rating Scale or CAPP IRS. The name was recently changed to better reflect the nature and intended uses of the test in community as well as institutional environments.
a)UK Research
•The CAPP IRS was applied to a sample of 72 male forensic patients with a primary diagnosis of schizophrenia spectrum disorders. It was found that in this sample the CAPP IRS had closer association with clinical features. Moreover, there appeared to be a larger overlap between CAPP IRS and schizophrenia symptoms than there was with the PCL R (De Page, Mercenier and Titeca, 2018).
RATED page updated: August 2019 © Risk Management Authority 2019
It has been used with individuals with learning disabilities in practice settings (Cooke, 2019, personal communication).
Other Considerations
•The CAPP SRS could be useful when measuring changes in the severity of symptoms over time. The fact that it can measure change makes it appropriate for risk management and is generally more acceptable to clients than other measures of PPD.
•Dawson and colleagues (2012) found that there were both strengths and challenges to using the CAPP IRS. The format of rating items on the CAPP IRS requires assessors to gather and consider a broader range of information. This is beneficial in providing a more comprehensive overview of cases; however, it also increases the time and effort involved in an assessment.
Contribution to Risk Practice
•An advantage of the CAPP IRS is that it covers a wider range of symptoms than other tools intended for the assessment of psychopathy (e.g. PCL:YV). Further to this, it allows for symptoms to be broken down into their component parts to allow for greater exploration of their nuances (Dawson et al., 2012; Kreis and Cooke, 2012; McCuish et al., 2019).
Applicability: Mental Disorders
•The CAPP SRS supports the proper assessment of PPD. It provides both the structure and process for carrying out one of the most challenging tasks in forensic practice (Cooke and Logan, 2018).
No empirical research at present.
b)International Research
•The motivation for constructing the CAPP conceptual model was to aid clinical evaluation through the development of new measures of PPD symptoms. The CAPP concept map has been translated into more than 25 languages. The CAPP SRS has been field tested in a range of settings (prisons, hospitals and secure units) in a number of countries (e.g., UK, Spain, Denmark, Belgium, Norway and Korea) (Cooke, 2018).
•The timeframe for using the CAPP SRS is flexible, ranging from short term (3, 6 or 12 months) or longer term (2 or 5 years or even lifetime) (Cooke et al., under review).
•In a study with 87 officers who rate psychopathy, it was found that the majority of CAPP symptoms (28 out of 33) were rated as highly or moderately typical of psychopathy. There remaining five symptoms were rated by practitioners as not typical of psychopathy: lacks pleasure, lacks perservance, lacks concentration, unstable self concept and lacks planfulness (Pauli et al., 2018).
•For more information, visit the following website: http://capp network.no
•A study by Kreis and colleagues (2012) employed 132 international mental health professionals to rate the symptoms of the CAPP in terms of their representativeness of psychopathy. The content validity of the CAPP was found to be good, with the majority of symptoms being highly representative of psychopathy in sensitivity and specificity. The items with the lowest prototypicality ratings were lacks concentration, lacks pleasure and unstable self concept.
•Convergent validity is evident between the PCL R and the CAPP IRS (now CAPP SRS), supporting that they assess the same underlying psychopathy construct (Sandvik et al., 2012).
•Practitioners should note that this is a clinical tool that assesses the construct of PPD and is therefore not a risk assessment instrument. It assesses constructs that have relevance for risk formulation and risk management.
•The CAPP is potentially useful in a variety of settings (e.g. correctional, forensic psychiatric, civil psychiatric, community and family ), rather than being optimised for use in a single setting.
•The CAPP comprises a family of tests. The current version, the Symptom Rating Scale (CAPP SRS) is designed for use in secure treatment settings (e.g. forensic psychiatric hospital).
•It is recommended that interviewers are initially paired to accommodate the extent of information collected as part of a CAPP assessment. It is also recommended that interviewers consistently debrief each other after interviews to ensure the correct information is gathered (McCormick, 2004)
RATED page updated: August 2019 © Risk Management Authority 2019
•The family of instruments will include; (1) Informant Rating Scale and (2) Clinical Interview.
•Construct validity for the CAPP SRS was found to be good, with it discriminating between three psychopathic traits without relying on the assessment of criminal behaviour (Florez et al., 2018).
•The internal consistency of the CAPP IRS (now CAPP SRS) was found to be good, except for the Cognition and Emotional Domains (De Page, Mercenier and Titeca, 2018).
•The CAPP SRS assessment is currently under going validation in many countries.
Author / Publisher Loranger Year 1997
Description
•The practitioner may adapt questions to suit the interview. The advantage of semi structured interviews is they incorporate the standardisation of a structured interview with flexibility to allow
Category Responsivity Issues (Validated)
•Symptoms must be present for at least five years. It is not appropriate for clients with severe depression, psychosis, low intelligence or cognitive impairment. With individuals in remission from chronic mental illness, discretion is advised on behalf of the user.
It is further recommended that at least one criterion of a disorder must have been fulfilled prior to age 25 before that particular disorder can be diagnosed.
Name of Tool International Personality Disorder Examination (IPDE)
Age Appropriateness
Whilst the IPDE is not suitable for those aged under 18 years, the manual starts that some investigators following slight modifications have found the tool useful for those as young as 15 Theyears.authors
•It involves a semi structured clinical interview developed to assess personality disorders as defined in the DSM IV and ICD 10. It also contains a self administered screening questionnaire.
recommend that for optimal usage clients should be aged 20+. The authors discourage the use of anything less than a five year timeframe with individuals over 20 years of age.
•It is useful for multiple professions and is based on worldwide field trials.
RATED page updated: August 2019 © Risk Management Authority 2019
•The IPDE is a semi structured clinical interview designed to assess the personality disorders in the ICD 10 and DSM IV classification systems. The IPDE ratings are current and, therefore, sensitive to change.
•It is possible to assess personality disorders with reasonably good reliability in different nations, languages and cultures (Loranger et al., 1994)
18 70 years
This is intended for use with experienced psychiatrists, clinical psychologists and those with comparable training. Users should have knowledge of the ICD 10 and DSM IV personality disorder criteria and experience of making psychiatric diagnoses.
Strengths
Assessor Qualifications
•The IPDE was administered by 58 psychiatrists and clinical psychologists to 716 patients enrolled in clinical facilities in Austria, Germany, India, Japan, Kenya, Luxembourg, the Netherlands, Switzerland and the United States. Thirteen percent of items yielded IRRs of 0.9 1; 72% and 52% rated 0.8 0.89 and 0.7 0.79 for IRR respectively. The author reports "inter rater reliability and temporal stability that is roughly similar to instruments used to diagnose psychoses, mood, anxiety and substance use disorders" (Loranger, Janca and Sartorius, 1997: 90).
It is based on personality traits prominent in the international field. It uses the general principles of personality disorder assessment (Cooke and Hart, 2000): disturbances in behaviours and personal relationships, as well as construct validation. The personality disorder constructs primarily reflect the views of Western European and North American psychiatry; thus, many no be equally applicable in other cultures (Loranger et al., 1994). It incorporates DSM IV and ICD 10 personality disorder evaluations. The DSM IV is predominantly used in the USA and the ICD 10 is mainly international psychiatric community opinion.
To that end, semi structured interviews provide a degree of procedural validity that makes their results more transferable and less susceptible to institutional and regional biases (Loranger, Janca and Sartorius, 1997).
•Dimensional scores are provided for every individual for each disorder, even in cases where they do not fulfil the criteria. The dimensional scores provide investigators with greater reliability and more versatility in data analyses (Loranger, Janca and Sartorius, 1997).
b)International Research
Empirical Grounding
Inter Rater Reliability
a)UK Research
Validation History
•Two UK sites (London & Nottingham) were included in the field trials for the IPDE. The overall IRR across all sites rated highly with IRR of 0.9 1, 0.8 0.89 and 0.7 0.79 respectively in 13%, 72% and 52% of items (Loranger, Janca and Sartorius, 1997).
•In a study in Indian, a Hindu translation of the IPDE was tested for its inter rater reliability. ICC ranged from 0.65 1.00 (m=0.89) for each item and between 0.94 1.00 (m=0.98) for dimensional score for each personality disorder (Sharan et al., 2002).
the interviewer to build rapport and ensure the interview flows Kvale & Brinkmann, 2009: 16, 27).
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
Applicability: Mental Disorders
•The DSM IV and ICD 10 IPDE SQ screeners were used and compared with the diagnoses obtained with the IPDE semi structured interview in a sample of 125 adolescents treated in a psychiatric department. The aim of the study was to analyse the usefulness of the IPDE Screening
a)UK Research
b)International Research
•Study groups internationally (including African, North American, European and Asian countries) included a mixture of male and female patients. There were no differences between urban and rural samples, or between men and women. (Loranger, Janca and Sartorius, 1997)
•Evidence of London (Maudsley Hospital) and Nottingham (Stonebridge Research Centre) involvement in clinical trials (Janca and Pull, 1997).
•The field trials held in the UK included an almost equal number of and female patients. Nottingham had 26 males and 24 females; whilst London had 34 males and 31 females. No differences were found between the genders (Loranger, Janca and Sartorius, 1997).
Validation History
None available at present.
b)International Research
Applicability: Females
General Predictive Accuracy
The IPDE is not a predictive instrument. As stated in the IPDE Manual: "The IPDE is a Semi structured clinical interview developed within that program and designed to assess the personality disorders in ICD 10 and DSM IV classifications systems.”
a)UK Research
b)International Research
Applicability: Ethnic Minorities
•A range of international studies indicate applicability to a range of ethnic groups. (El Rufaie, 2002; Fountoulakis, 2002; Magallon Neria et al., 2012; Mann et al., 1999; Sharan et al., 2002).
a)UK Research
Validation History
RATED page updated: August 2019 © Risk Management Authority 2019
•Provides an individualised process for identifying the presence of a personality disorder.
Questionnaire (IPDE SQ) for identifying DSM IV and ICD 10 Borderline and Impulsive personality disorders in Spanish adolescents. The cut off point with the best combination of sensitivity and specificity for ICD 10 borderline and impulsive personality disorders was obtained with 3 positive items (Magallón Neria et al .,2012).
•The Greek translation of the IPDE has also shown applications whilst being mindful of cultural variation around socio cultural factors (Fountoulakis, 2002).
•Ninety psychiatric out patients in India, were assessed for personality disorder using the IPDE and Standard Assessment of Personality (SAP) methods. The overall agreement between the two instruments in the detection of ICD 10 personality disorder was modest (kappa = 0.4). The level of agreement varied according to personality category, ranging from kappa 0.66 (dependent) to kappa 0.09 (dissocial) (Mann et al., 1999)
•A sample (n = 158) of primary health care patients in United Arab Emirates (UAE) were interviewed by general practitioners (GPs) using the Arabic version of the IPDE (ICD 10 version). This was useful but relatively time consuming with repetition and need of rephrasing in some items. Dimensional measurement proved essential (El Rufaie et al., 2002).
•Interview stage should be between an hour to ninety minutes. The validation study did find there was considerable variation amongst interviewers about this figure, with the average length of an interview being cited as around 2 hours and 20 minutes. The authors caution that if an interview were to exceed more than an hour and a half, there is a risk that the assessor will not pursue responses with the same degree of alertness and thoughtfulness and/or the individual’s replies will become briefer and more perfunctory in nature. It is recommended that in those situations, the interview should be continued over several stages; although it is best to avoid interrupting an interview in the middle of a section (Loranger et al., 1994).
RATED page updated: August 2019 © Risk Management Authority 2019
•The dimensional scoring can inform a formulation based approach to risk assessment by identifying the presence of specific traits.
Contribution to Risk Practice
•The IPDE SQ is intended to be an initial screen to detect likely personality disorder to then be followed by a comprehensive assessment. It is a self administrated form consisting of 57 items written at a nine years of age reading level, which can be completed in fifteen minutes or less (Slade
Other Considerations
•The IPDE allows for the use of assessor ratings and additional information that could be relevant.
•It is acknowledged that there may be an impact on the continued use of the DSM IV version of the IPDE with the arrival of DSM V. The ICD 10 version of IPDE remains relevant.
•The IPDE ratings should be based on life long patterns and the typical functioning of an individual.
•Slade and Forrester (2013) recommend the cut off score is adjusted for certain populations. The standard for the IPDE SQ is three affirmative answers; however, there are validity issues with this in certain populations: prisoners, adults seeking speech treatment for stuttering and smokers).
and Forrester, 2013). This is useful for identifying those who would be unlikely to meet the criteria for diagnosed personality disorder, but has the tendency to produce high numbers of false positives.
RATED page updated: August 2019 © Risk Management Authority 2019
Increasing the cut off to four or more answers is reported to be a suitable validity index for these populations. For instance, in the case of prisoners, increasing the cut off score accounts for some aspects of prison culture (e.g. fear, anxiety) to be accounted for.
Description
•The PCL R produces dimensional scores; it can also be used to classify or diagnose individuals for research and clinical purposes.
Author / Publisher Hare Year 2003
18+ Assessor Qualifications
•Large research base for the PCL R. It measures personality traits and behaviours relating to a widely understood concept of psychopathy (e.g. Berrios, 1996; Cleckley, 1976; Pichot, 1978).
Clinicians should possess the following qualifications: (1) an advanced degree in the social, medical or behavioural sciences; (2) completed graduate courses in psychopathology, psychometric theory and statistics; (3) knowledge of the clinical and research literature relating to psychopathy; (4) professional credentials with the appropriate regulatory body that regulates the assessment and diagnosis of mental disorders or be legally authorised to conduct psychological assessments; (5) demonstrated experience with forensic or other relevant populations; (6) adequate training and experience in administering the PCL R.
Training workshops are offered by Professor Hare’s Darkstone Research Group, Ltd. (see http://www.hare.org/training/ for details) amongst other providers There is a web based training programme offered by the Global Institute of Forensic Research that has been certified by the Darkstone Research Group (see https://www.gifrinc.com/pcl r/ for details).
Strengths
•The PCL R is a 20 item scale for the assessment of psychopathy in research, clinical and forensic settings. It involves a semi structured interview, file and collateral information.
Category Responsivity Issues (Validated)
•The PCL R has a categorical use, whereby its scores indicate whether an individual meets the criteria of a psychopathy. It also has a dimensional use, relating to interpersonal or affective (Factor 1)or behavioural (Factor 2) features of psychopathy (DeMatteo and Edens, 2006).
RATED page updated: August 2019 © Risk Management Authority 2019
Age Appropriateness
Name of Tool Psychopathy Checklist Revised (PCL R)
•Examiners rate each item on a 3 point scale: 0 (not applicable the individual does not exhibit the trait or behaviour in question); 1 (applies to a certain extent a match in some respects but with too many exceptions or doubts); 2 (applies a reasonably good match in most essential respects).
Since 1980, the PCL R has been developed and found to be applicable to diverse populations and identifies psychopathy as a risk factor for violence in both mentally and non mentally disordered PCLindividuals.Rscores have been incorporated into other instruments such as the VRAG and the HCR 20 (Hare, 1991). These are no longer included in the revised version of the VRAG and the third version of the HCR 20.
RATED page updated: August 2019 © Risk Management Authority 2019
•For a measure that is not a risk assessment, it has the ability to predict violent recidivism (Daffern, 2007).
a)UK Research
•Morrissey et al. (2007) the PCL R achieved high ICC of .80 within high secure forensic settings.
•Blais, Forth and Hare (2017) carried out an examination of inter rater reliability across a sample of 280 trained raters. It was found that the cases of individuals with high psychopathy scores showed better reliability than those with moderate to low ones. A public significance statement was released with the article cautioning that whilst the reliability of the PCL R was good amongst the raters attending the training, it did not meet the recommended standard for criminal cases.
•Rettenberger et al. (2010) found an excellent ICC value of .93 for the PCL R
•Laurell and Daderman (2007) reported an excellent ICC value for the PCL R (.96)
b)International Research
•Hare (2003) reports ICCs of .86 for North American males who have offended.
Empirical Grounding
•Logan and Blackburn (2009) large correlation coefficients (rho) observed; between raters (r= .83), composite score (r= .73), Factor 1 scores (r= .73) and Factor 2 scores (r= .77).
Inter Rater Reliability
•Ismail and Looman (2016) examined the inter rater reliability for each of the PCL R scores using archival data of 178 sexually offending individuals based in a correctional facility in Canada. The ICC range was good to excellent for the individual score items, apart from pathological lying.
•Farrington, Jolliffe and Johnstone (2008) In a meta analytic study, the PCL R generated a moderate AUC value of .69.
•Coid et al. (2009) the PCL R generated moderate AUCs in the prediction of violence (.64) and general recidivism (.65).
•Olver and Wong (2006) Composite PCL R score obtained moderate predictive accuracy in relation to re offending rates for non sexual and sexual offending (AUCS= .61 .73)
a)UK Research
•Abbiati and colleagues (2018) applied the PCL R to 52 individuals with violent offences in a Swiss prison to evaluate its predictive validity for different types of misconduct. Fair predictive validity was shown for physically violent misconduct with an AUC of 0.78; poor predictive validity was shown for any misconduct and other misconduct (AUCs of 0.65 and 0.66 respectively).
RATED page updated: August 2019 © Risk Management Authority 2019
•Olver et al. (2013) examined the PCL R in large samples of Canadian Aboriginal and non Aboriginal individuals Analyses of predictive accuracy found that medium effects were found in predicting violent, non violent and general criminal recidivism for both groups.
Validation History
General Predictive Accuracy
•Cooke et al. (2001) the PCL R generated a moderate AUC score of .65.
•Krstic et al. (2017) administered structural equation modelling and found that the PCL R factors provided a basis for allocating those who have committed sexual offences into four distinct sub types.
•Coid et al. (2007) the PCL R had moderate predictive accuracy for various types of offences: Violence (AUC=.64), Theft (AUC=.66), Drugs (AUC=.60), and Any Offence (AUC=.65).
•Rettenberger et al. (2010) found the PCL R generated high AUC values for sexual recidivism (.73), general violent recidivism (.75), and general criminal recidivism (.80) in a sub group of sexual offending individuals.
b)International Research
Applicability: Females
•Hawes, Boccaccini and Murrie (2013) carried out a meta analysis looking at the relation between sexual recidivism (combined sample size of 5239) and PCL R scores. The total score effect was d=0.40, which is at the upper end of confidence intervals. These effects were stronger against for Factor 4 (d=0.40) and Factor 2 (d=0.44). Moreover, effect sizes tended to be stronger for scores calculated for research purposes (d=0.44) not clinical use (d=0.28).
•Vitale et al. (2002) found small to large Pearson correlations between the PCL: R composite score and violent and non violent offending which ranged from .18 to .44.
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
•DeMatteo et al. (2014) carried out a review of the PCL R in 214 cases of sexually violent predator offending in the U.S. There were multiple scores in 29 of the cases and an ICC of only .58, suggesting that the PCL R may not be a suitable tool for these types of cases.
•Schaap, Lammers and de Vogel (2009) found above chance AUC values for violent recidivism (.57) and moderate AUC for general recidivism (.60).
a)UK Research
•A study examined 108 clinicians’ scoring of the PCL R using case materials and a seven point scale to provide a rating of an individual’s risk of committing a new sexual offence. Results showed that their judgments were more strongly associated with assigned Factor 1 scores than Factor 2. This is in spite of the fact that Factor 1 traits not necessarily being the most predictive of future risk; although they are the most prototypical of psychopathy (Gardner, Boccaccini and Murrie, 2018).
b)International Research
•Gray and Snowden (2016) examined psychopathy in female psychiatric patients in the UK and the US using the PCL:SV. Based on their findings and other studies, the authors surmised that the PCL R and the PCL:SV are predictive of antisocial outcomes in women and that there is very little difference when compared to findings involving male patients.
None available at present.
•Olver et al. (2018) carried out a study on Canadian indigenous and nonindigenous males. Findings indicated that indigenous men scored higher on most components of the PCL R and had higher rates of recidivism than nonindigenous males.
•In a sample of 78 female forensic patients, the PCL R demonstrated good predictive accuracy for all recidivism within a 3 year follow up period (AUC=.710); over a longer period of time the AUC for all recidivism dropped to .60. Violent recidivism generated low predictive accuracy with an AUC of .457. the authors postulate this may be attributed to female psychopaths engaging in subtle, manipulative rather than violent behaviour (de Vogel, Bruggeman and Lancel, 2019).
None available at present.
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
b)International Research
Applicability: Ethnic Minorities
•Skeem et al. (2004) meta analysis suggested no strong evidence of differences in the core psychopathic traits in White and Black participants.
Validation History
•In the Weizmann Henelius et al. (2010) study, raters used forensic examination reports to use the PCL R to retrospectively rate Finish females who have offended. A robust association was found between psychopathy and borderline personality disorder. It was also discovered that the impulsive and unstable features of psychopathy have a greater presence in females with the disorder, suggesting psychopathy may be expressed differently in men and women.
Validation History
•Sullivan et al. (2006) moderate correlations observed between the composite PCL: R scores of ethnic minority indivdiuals and violent and non violent behaviours.
•Tsang, Piquero and Caufman (2014) applied the PCL:YV to male adolescents of Caucasian, African American and Hispanic ethnicities. It was found that there was substantial, differential item functioning in 15 of the 20 items across the ethnic groups.
•Although not designed primarily to identify factors associated with further offending, it reviews factors that are established as general risk factors, and others that would be relevant to risk management planning.
•Morrisey et al. (2010) examined the use with Intellectually Disabled individuals finding a preference for using the instrument for clinical purposes as opposed to focusing on total scores.
•The PCL R can aid assessors in identifying risk and responsivity factors specific to the individual, such as a lack of remorse or guilt and failure to accept responsibility for own actions.
•McDermott et al. (2008) composite PCL:R scores did not significantly predict inpatient violence (AUC= .58). Factor 2 scores obtained moderate predictive accuracy in ‘Aggression towards Staff’ (AUC=.66) and ‘Aggression towards Patients’ (AUC=.65).
•Assessments using the PCL R have been used in a variety of criminal justice settings throughout Western society: civil commitment proceedings in the U.S.; dangerous offender hearings in Canada; severe dangerous personality disorders in the UK. In the Netherlands, it is also a requirement that the PCL R is administered to all forensic psychiatric inpatients.
Contribution to Risk Practice
b)International Research
a)UK Research
•Some PCL R items can be targets for change.
•While in some contexts a categorical conclusion about psychopathy is required, the cut off score is primarily used to facilitate comparative research.
•The PCL R is a four factor model identifying the traits related to the construct of psychopathy: interpersonal/affective features (e.g. callousness and superficial charm); lifestyle (e.g. irresponsibility and impulsivity); antisocial (e.g. poor behavioural controls and early behavioural problems).
•Where an assessor is required to offer a categorical conclusion and report a cut off score, it is important that s/he is aware of the relevant research and uses the most appropriate normative data for the population. There is evidence to suggest cross cultural validity (Hare, 1998) and variability and accuracy of cross cultural cut off scores (Cooke and Michie, 1999). This is particularly
Applicability: Mental Disorders
•The PCL R has been designed to assess the presence of psychopathic traits rather than the risk of recidivism (albeit that the presence of psychopathy has been shown in prior investigations to be a risk and responsivity factor for recidivism and response to treatment/intervention respectively).
•Psychopathy, as measured by the PCL R and its screening version PCL:SV, is part of other risk assessment tools: VRAG, SORAG, DVRAG and SVR 20. The HCR 20 includes information drawn from PCL R assessments (Douglas and Reeves, 2010; Hare, 2003; Hare and Neumann, 2009).
•The manual notes that for clinical assessments a dimensional approach to interpreting the findings of a PCL R is often preferred. In this approach an assessor may use the PCL R to identify the presence of psychopathic traits; to consider their relevance to risk management; compare an individual’s total and factor scores against percentiles.
•Logan and Blackburn (2009) moderate correlations observed for non violent convictions and factor 1 (r= .29) and factor 2 (r =.27) scores in high secure settings.
RATED page updated: August 2019 © Risk Management Authority 2019
•Assessing 269 young males who committed violence offences, González and colleagues (2019) found that there was a positive association between psychopathic traits and aggressive antisocial behaviours, with the strongest correlations between the Lifestyle and Antisocial facets of the PCL R. Additionally, there was no association between the verbal dimensions of intelligence and PCL R facets, suggesting that even though persons with psychopathic traits may seem to demonstrate an above average intelligence level this may be misleading.
•Some research suggests that a diagnosis of psychopathy could be regarded as a negative label with negative consequences on sentencing, treatment and clinical judgement (Lloyd, Clark and Forth, 2010). Other investigations suggest that a defendant’s prior criminal history holds more influence over sentencing than a diagnosis of psychopathy (Cox et al., 2010).
RATED page updated: August 2019 © Risk Management Authority 2019
•For more information, please visit the following website: www.hare.org
Other Considerations
•There is the absence of a clear cut cut off score for diagnosing an individual as a ‘psychopath.’ Generally, groupings around the possible presence of the disorder are informed by the following scores: low for a total of below 20; medium for scorings between 20 30; scores of 30 and more as high.
•De Matteo, Edens and Hart (2010) recommended that the PCL R should be used as part of a comprehensive risk assessment investigation; rather than it being the sole measure.
•Psychopathy as a construct is relatively stable, therefore it cannot account for fluctuations in mental states or behavioural change (Daffern, 2007). Psychopathy can encompass traits from other personality disorders: anti social, narcissistic, histrionic and paranoid (NHS England and National Offender Management Service, 2015: 141).
•Items should be omitted only when absolutely necessary (i.e. there is insufficient information to correctly score an item). The omission of too many items will deplete the reliability of the PCL R. Also, there are no provisions in place to allow the user to modify or veto an item score. The authors, therefore, advise that the PCL R is used strictly as designed or not at all.
•PCL R items are rated on the basis of lifetime functioning. Socio demographic factors like race and class may influence the meanings of items as well as the practical implications of an assigned score.
relevant within a forensic population where it’s suggested that a cut off of 25 is more accurate for England and Scotland (Cooke and Michie, 1999).
•Based on empirical research carried out, Morrissey (2013) has produced a set of guidelines to be used as a supplement to the PCL R and the PCL:SV manuals with males with intellectual disabilities. For instance, it is documented that interview evidence may be less reliable in individuals with IDs and to assist this process there should be increased time for interviewing and the standard questions should be adapted. It is also recommended that the PCL R should not be used in individuals with IDs under 21 years old, due to developmental differences.
•Assessors should note that this tool has been normed on forensic mental health samples, however, in certain sub groups of mentally disordered individuals (e.g. learning disabilities) its accuracy in predicting recidivism lessens.
•Acknowledging the differences of original and local validation samples assessors should ensure they rate each item carefully and then examine the cultural and social context in which the assessment was made in order to determine and understand differences. (e.g. differences between U.K. and North American subjects in the interpersonal style items, particularly grandiose sense of self worth and glib/superficial charm (Cooke et al., 2005).
•The PCL:SV omits items scored on the basis of them being challenging to confirm or too detailed. This means that an interview using the PCL:SV can be completed in around 30 to 60 minutes.
16+ is prescribed in the manual (Hare, 1995).
•The tool is deemed as highly reliable when used by individuals with the relevant experience and training.
Age Appropriateness
•The PCL:SV is a 12 item abbreviated tool derived from the PCL R designed to screen for the possible presence of psychopathy.
Year 1995
Assessor Qualifications
It is, however, recommended by Multi Health Systems, Inc. that the PCL:SV is used with individuals aged 18 and above. This makes sense considering that PCL R cannot be administered to a 16 or 17 year old who was demonstrating a high score on the PCL:SV.
RATED page updated: August 2019 © Risk Management Authority 2019
Description
•Cut off scores indicate when to follow up with the full PCL R assessment.
•The tool was not designed to replace the PCL R but to offer an efficient tool to screen for the possible presence of psychopathy in those who have offended and forensic psychiatric patients (Hart, Cox and Hare, 1995).
Category Responsivity Issues (Validated)
Name of Tool Psychopathy Checklist: Screening Version (PCL:SV)
Author / Publisher Hart, Cox and Hare
Strengths
Similar specifications as with its predecessor, the PCL:R.
•Criminal records are not needed for this tool, making it more appropriate than the PCL R for use in non forensic as well as non criminal settings. The authors maintain it is particularly suitable for civil psychiatric evaluations, personnel selection in law enforcement and the military, and studies of community residents.
•The tool is widely used in non forensic contexts, both as a screen for psychopathy and as a ‘stand alone’ instrument, particularly with community and psychiatric populations (Guy and Douglas, 2006; Oliveira Souza et al., 2008), particularly in countries outside of North America (Douglas et al., 2005).
•Doyle et al. (2012) found high ICCs of .97 for the composite score, .85 for factor 1 scores and .80 for factor 2 scores.
b)International Research
Empirical Grounding
•Campbell, French and Gendreau (2009) meta analytic research on a variety of risk assessments revealed that the PCL:SV produced the third largest mean effect size (N = 504, K = 7, Z+ = .22) in predicting institutional violence and a strong magnitude for predicting violent recidivism (K= 5, N =641, Z+ .20).
Inter Rater Reliability
•Campbell, French and Gendreau (2009) meta analytic research on a variety of risk assessments revealed that the PCL:SV produced the third largest mean effect size (N = 504, K = 7, Z+ = .22) in predicting institutional violence and a strong magnitude for predicting violent recidivism (K= 5, N =641, Z+ .20).
a)UK Research
•The tool is a derivative of the PCL:R and is conceptually and empirically related to the PCL:R (Guy and Douglas, 2006).
b)International Research
•Gray et al. (2004) also found large correlations for the PCL:SV composite score (r= .98).
a)UK Research
•Žukauskienė, Laurinavičius and Čėsnienė (2010) the PCL:SV composite scores obtained moderate correlations in relation to criminal convictions (r=.26), violent offending (r= .22) and total time spent in correctional institutions (r=.20).
Validation History
RATED page updated: August 2019 © Risk Management Authority 2019
•The tool correlates approximately with the longer version in the normative sample (.80) (Hart, Cox and Hare, 1995).
•Dietiker, Dittmann and Graf (2007) compared the PCL:SV, HCR 20 and SVR 20 in a German sample of individuals with sexual offences and confirmed the utility of PCL:SV.
•Howard (2007) the PCL:SV was found to be a moderate predictor of future violence (AUC = .64) in a sample of individuals serving community sentences.
General Predictive Accuracy
•Richards, Casey and Lucente (2003) scores on the PCL:R and PCL:SV were significantly associated with disruptive and violent rule violations and other non compliant behaviours.
•Morrisey et al. (2010) examined the use of the PCL:SV with Intellectually Disabled individuals, finding a preference for using the instrument for clinical purposes as opposed to focusing on total scores.
Validation History
None available at present.
Validation History
•Dietiker, Dittmann and Graf (2007) compared the PCL:SV, HCR 20 and SVR 20 in a German sample of individuals with sexual offences and confirmed the utility of PCL:SV.
•Žukauskienė, Laurinavičius and Čėsnienė (2010) PCL:SV composite scores obtained moderate correlations in relation to criminal convictions (r=.26), violent offending (r= .22) and total time spent in correctional institutions (r=.20).
b)International Research
a)UK Research
Applicability: Mental Disorders
•Higgs, Tully and Browne (2018) found that the PCL:SV showed similar predictive accuracy to the PCL R with regards to violence risk.
Applicability: Ethnic Minorities
•Cullen et al. (2011) mentally disordered individual with scores of 16 and above on the PCL:SV were just over 13 times more likely to drop out of an offending treatment program than those with lower scores.
a)UK Research
RATED page updated: August 2019 © Risk Management Authority 2019
Validation History
No empirical evidence at present.
Applicability: Females
b)International Research
•Ho, Thomson and Darjee (2009) ROC analyses revealed that the PCL:SV had moderate predictive accuracy for predicting serious violence (AUC = .66) and any violent incidents (AUC = .63) in a sample of mentally disordered individuals.
Contribution to Risk Practice
•High scoring on the PCL:SV can be indicative of a need to administer the PCL R tool. The PCL:SV has moderate false positive (i.e. an individual wrongly being categorised as a psychopath) and very low false negative rates (i.e. an individual who meets the criteria of a psychopath not being recognised as one).
•A score of 18 and above is generally used as a marker for psychopathy; scoring of 12 and lower is considered to be achieved only with non psychopaths.
•Being a derivative of the PCL:R , the PCL:SV will also suffer similar disadvantages.
RATED page updated: August 2019 © Risk Management Authority 2019
•Guidelines were produced by Morrissey (2013) about using this tool with individuals with intellectual disabilities. It was recommended that due to the developmental delays in an individual with intellectual disabilities, the PCL:SV should not be used with those aged under 21 years with IDs.
•Similar to its ‘parent’ tool, the PCL R, the PCL:SV should be used to test the lifetime functioning of an individual; it should not be used for assessments pertaining to the ‘present state’ or a brief period of time (less than a year). It is also not designed to identify risk factors; rather, it reviews factors that would be relevant to risk management planning.
•Arbach Lucioni et al. (2011) the PCL:SV displayed moderate predictive accuracy in predicting inpatient violence in the short term (AUC=.70) however its accuracy lessened in the follow up periods (AUC=.61).
•Gray et al. (2004, 2007) moderate to large AUCs found for recidivism in a sample of those with mental disorders.
Other Considerations
•The PCL:SV provides a brief scan of factors related to the construct of psychopathy some of the factors analyse the individual’s past and current offending behaviours. This information can be used to prompt further assessment of identified risk factors. A study by Stoll and colleagues (2019), for instance, found there was low levels of psychopathy in a sample of low risk individuals who had committed child sexual offences (43 paedophilic offences and 21 were control participants); although a higher level of neuroticism was associated with higher PCL:SV scores.
•Douglas et al. (2005) in a sample of male and female forensic psychiatric patients, composite PCL:SV scores were moderately predictive of inpatient aggression (AUCs= .63 .68).
RATED page updated: August 2019 © Risk Management Authority 2019
•For more information, please visit the following website: www.hare.org .
•There is debate within research regarding the potential consequences of a diagnosis of psychopathy and its effects on sentencing, treatment and clinical judgement (Cox, DeMatteo and Foster, 2010; Lloyd, Clark and Forth, 2010).
•As noted under the PCL:R, validation research relating to the PCL:SV should also be interpreted with caution given that the tool was designed to screen for psychopathic traits rather than assess the likelihood of recidivism.
•Unlike the PCL:R, the PCL:SV can be completed in the absence of criminal record information, which increases its versatility outside of forensic settings (Hart, Cox and Hare, 1995).
RATED page updated: August 2019 © Risk Management Authority 2019
Year 2003
Similar specifications as with its predecessor, the PCL:R: (1) advanced graduate degree in the social, medical or behavioural sciences; (2) possess appropriate professional credentials; (3) a familiarity with the clinical and research literature pertaining to psychopathy, both in adults and adolescents; (4) experience working with adolescents and/or familiarity with developmental norms; (5)adequate training and experience in using the PCL:YV; (6) avoid using non standard procedures to administer the PCL:YV. A qualified clinician should supervise assessors who do not have the above qualifications (Hare, 2003).
•The PCL:YV is a 20 item scale designed specifically for the assessment of psychopathic traits in adolescent populations (Brazil and Forth, 2016)
Category Responsivity Issues (Validated)
Individuals can administer the PCL:YV for clinical purposes if they are licensed to conduct psychological assessments and possess an advanced university degree (postgraduate level). Furthermore, it is recommended that assessors establish interrater agreement through training before using the instrument for clinical purposes. Use of the PCL:YV for research purposes only does not require assessors to be licensed professionals (Brazil and Forth, 2016).
Description
•The PCL:YV is suitable for both male and female populations between the ages of 12 and 18 (Forth, Kosson and Hare, 2003).
12 18
•The authors maintain that the PCL:YV is a ‘downward extension’ of the PCL R tailored to be more applicable to the target population. The PCL R items pertaining to adults such are ‘Parasitic lifestyle,’ ‘Lack of realistic long term plans’ and ‘Many short term martial relationships’ were replaced by items that attempt to capture similar dispositions but in the forms they appear during adolescence in the PCL:YV. The item descriptions and scoring guides for several other items were also modified. For example, ‘Juvenile delinquency’ and ‘Criminal versatility’ were also modified, given that adolescents have less contact with the justice system than adults at this stage in their lives (Forth, Kosson and Hare, 2003)
Author / Publisher Forth, Kosson and Hare
Assessor Qualifications
Name of Tool Psychopathy Checklist Youth Version (PCL: YV)
•Using a semi structured interview and collateral information, the PCL:YV measures interpersonal, antisocial, affective, and behavioural features related to a widely understood, traditional concept of psychopathy (Forth, Kosson and Hare, 2003)
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
This youth version is a downward extension of the PCL:R, used to assess personality characteristics and elements of psychopathic behaviour (Forth, Kosson and Hare, 2003).
•Cauffman et al. (2009) obtained excellent ICC value of .91 for the composite PCL:YV scores.
•Dolan and Rennie (2008) poor to moderate ROC values for general recidivism (.60) and violent recidivism (.54).
a)UK Research
•When used in appropriate contexts and by appropriately trained professionals, can be useful in directing future treatment and other interventions (Brazil and Forth, 2016).
•Dolan and Rennie (2006) measured the inter rater reliability of three researchers. ICCs ranged between .87 to .93.
•Marsh et al. (2011) excellent correlation coefficient found for PCL:YV scoring (r =.91).
Validation History
General Predictive Accuracy
Inter Rater Reliability
•Marshall et al. (2006) large AUCs for the PCL:YV across 3 offending categories: recorded incidents of violence (.73), number of charges and convictions (.73) and assaults (.75).
•Douglas, Epstein and Poythress (2008) moderate to large AUCs observed for violent (.66) and weapons related (.88) recidivism. PCL: YV scores did not, however, significantly predict any recidivism or non violent recidivism.
b)International Research
•Strong empirical grounding given that the measure draws upon the research of the PCL:R (Forth, Kosson and Hare, 2003).
Empirical Grounding
•Welsh et al. (2008) excellent ICC value of .84 obtained for the PCL:YV.
a)UK Research
•McCuish et al. (2019) excellent ICC was of .92 was found for the PCL:YV total score.
b)International Research
Strengths
•Shepherd and colleagues (2014) applied the PCL:YV to a sample of Australian young offenders over a period of up to 18 months in order to ensure a minimum follow up of six months. The PCL:YV was found to predict general and violent recidivism generating AUCs of .66 and .64 respectively.
•Edens and Cahill (2007) the PCL: YV did not significantly predict violent and general recidivism in community settings.
•Bauer, Whitman and Kosson (2011) Moderate to large correlations observed between total number of charges (r= .29), number of violent infractions (r= .38) and the total number of infractions (r= .43) in a sample of institutionalised female offenders.
•Stockdale, Olver and Wong (2010) moderate to large AUCs found for the original 20 item 4 factor model ranging from .67 to .68 for total recidivism and from .70 to .75 for youth recidivism. It was, however, unable to significantly predict adult recidivism.
•Marshall et al. (2006) satisfactory correlations found between the PCL:YV scores and recidivism in relation to predicted assaults, total charges and reconviction in a group of female offenders.
Applicability: Females
RATED page updated: August 2019 © Risk Management Authority 2019
•Catchpole and Gretton (2004) the PCL:YV obtained a large AUC value of .73 in predicting violent recidivism.
•A study of 72 juvenile sex offenders by Wijetunga et al. (2018) tested the predictive validity of the PCL:YV. The AUCs for general nonsexual, violent nonsexual and sexual recidivism were .63, .54 and .77 respectively. The PCL:YV was also found to strongly correlate with Scale P of the JSOAP II, a youth sexual violence tool with a scale intended to measure psychopathy.
Validation History
•Corrado et al. (2004) found small to moderate AUC values in predicting any, non violent and violent recidivism ranging from .58 .68.
a)UK Research
None available at present.
b)International Research
Other Considerations
b)International Research
Applicability: Mental Disorders
a)UK Research
•Schmidt et al. (2006) the PCL:YV attained moderate predictive accuracy for violent (.71) and general (.72) recidivism in a sample of juvenile offenders who were referred for mental health assessments.
RATED page updated: August 2019 © Risk Management Authority 2019
•Stockdale, Olver and Wong (2010) moderate to large AUCs found for total recidivism (range = .71 to .72) and youth recidivism (range = .73 .81). Moderate AUC of .63 was found for adult recidivism in a group of Aboriginal offenders.
Validation History
•The authors advise that the PCL:YV should not be the sole decision making measure used to assess risk of recidivism. Its standard error of measurement should also be considered to account for ‘false positive’ cases where item scores are part of an adolescent development process.
a)UK Research
•McCuish and colleagues (2018) tested the predictive validity of the PCL:YV across 137 indigenous and 312 White adjudicated youth. Support was evident for using the PCL:YV across both ethnic groups. The lifestyle and antisocial factors were more informative of recidivism outcomes than interpersonal and affective factors.
•The PCL:YV provides an assessment of factors related to the construct of psychopathy some of the factors analyse the individual’s past and current offending behaviours. This information can be used to prompt further assessment of identified risk factors and other relevant intervention options.
•Schmidt et al. (2006) the PCL:YV attained excellent predictive accuracy (ROC) for violent recidivism (.83) and general recidivism (.76). Its accuracy in predicting non violent recidivism was, however, below chance (.31).
None available at present.
Contribution to Risk Practice
None available at present.
Validation History
Applicability: Ethnic Minorities
b)International Research
•As noted under the PCL:R, validation research relating PCL:YV scores to recidivism should also be interpreted with caution given that the tool was designed to screen for psychopathic traits rather than assess the likelihood of recidivism.
•For more information, please visit the following website: www.hare.org
•Fewer validation studies conducted on UK populations. Pechorro et al. (2015) found the Portuguese version of the PCL:YV demonstrated promising psychometric properties with regards to the three factor model of youth psychopathy; although further validation work is still required.
•Hemphälä and colleagues (2015) found there was moderate to high rank order stability as indicated by correlated with PCL R ratings five years later (overall rs=.68 for males and .58 for females). Further, excellent intra individual stability was found with 87% and 86% of males and females respectively exhibiting no reliable changes in PCL scores.
RATED page updated: August 2019 © Risk Management Authority 2019
•It may be possible to complete the PCL:YV solely using information contained in file records in cases where information provided during the interview with the offender is of little use (Forth, Kosson and Hare, 2003). Such assessments are considered nonstandard assessments.
•Concerns regarding the application of an adult construct to an adolescent population. Controversy regarding the applicability of some psychopathic traits to children and adolescents (e.g. impulsivity, parasitic lifestyle) (Edens, Petrila and Buffington Vollum, 2001; Kotler and McMahon, 2010).
•Schmidt and colleagues (2006) state that caution should be applied when using this tool. The label of psychopathy could be regarded as pejorative and may have negative effects on treatment, legal sentencing and community supervision. It was also suggested that few studies have explored long term and developmental correlates of high scores on the PCL:YV.
•Some researchers argue that general characteristics of adolescence can be mistaken for psychopathic traits (Edens, 2001).
•A cut off score for clinical diagnosis is not provided, in line with the recommendations of practitioners not to diagnose personality disorders in adolescents.
•Dawson and colleagues (2012) found that the PCL:YV indicated the presence of serious psychopathy related personality disturbance when it was applied to two incarcerated youth. They suggested that measures like the CAPP IRS could complement the use of the PCL:YV.
Age Appropriateness
RATED page updated: August 2019 © Risk Management Authority 2019
•The AAA is an electronic tool able to be scored on the computer (Baron Cohen et al., 2005).
Category Responsivity Issues (Awaiting Validation)
•In an effort to avoid ‘false positives,’ the CLASS criteria are more stringent and conservative than the internationally recommended guidelines in DSM IV. Anyone meeting CLASS criteria would also meet DSM IV criteria (Baron Cohen et al., 2005).
Name of Tool Adult Asperger Assessment (AAA)
•In order to deal with the number of adults with suspected Asperger syndrome, Baron Cohen and colleagues (2005) developed the AAA instrument. The AAA links with two screening instruments: the Autism Spectrum Quotient (AQ) and the Empathy Quotient (EQ). The AQ is a 50 item screen identifying core autistic features in adults, across five different areas: social skill, attention switching, attention to detail, communication and imagination. The EQ consists of 60 items, 40 of which signify the degree of empathy alongside 20 filler questions. Individuals must complete these screening tests before then attending an interview carried out by a clinician (Baron Cohen et al., 2005; Stoesz et al., 2011).
Assessor Qualifications
18+
Tool Development
•The AAA was developed in the CLASS (Cambridge Lifespan Asperger Syndrome Service) clinic, which provides a specialist diagnostic evaluation.
•Asperger’s Syndrome is a ‘pervasive development disorder’ located at the high end of functioning of the autism spectrum. It involves significant difficulties in social interaction in addition to restricted and repetitive patterns of behaviours and interests (Fabian, 2011; Murphy, 2007).
Author / Publisher Baron Cohen and Wheelwright updated by Bradley Year 2000, 2011
•Kenny and Stansfield (2016) examined AAA results in adults with intellectual disabilities diagnosed with Asperger’s syndrome. It was found that this population scored lower on the autism spectrum quotient and higher on the empathy quotient than those without intellectual disabilities.
A clinician with the relevant experience is required to carry out an interview with the patient.
Description
During the interview a check for the presence of symptoms relevant to a diagnosis of Asperger Syndrome (AS) or High Functioning Autism (HFA) is conducted.
•Baron Cohen and colleagues (2005) applied both the AAA and the DSM IV to 2 clinic patients. It was found that whilst 88% met DSM IV criteria, only 82% met AAA criteria. Based on this, the authors recommend that clinicians adopt the AAA to maintain a stricter definition of Asperger Syndrome.
•Females, children and adults with considerable intellectual ability may be more difficult to diagnose with Asperger’s disorder, as they tend to have greater abilities to hide their problems (Attwood, 2006).
RATED page updated: August 2019 © Risk Management Authority 2019
•The AAA utilises self reporting as part of the assessment, where the individual completes the AQ and EQ components of the instrument before attending a diagnostic interview with a clinician (Stoesz et al., 2011).
•Individuals with Asperger’s disorder tend to have problems with social interactions and understanding the emotions, reactions and experiences of other people (Barry Walsh and Mullen, 2004).
•Murphy (2007) looked at the PCL R profiles of 13 male patients with Asperger’s syndrome in a high security psychiatric care facility. It was found that those patients appeared to rate more highly on PCL R items relating to interpersonal and affective features rather than social deviance.
General Notes
•Another feature of the disorder may be abnormal, repetitive, narrow interests that translate into repetitive, focused and persistent behaviours. These features may lend themselves to criminal behaviour pertaining to the individual’s’ narrow interests (e.g. stalking, stealing and hoarding) and feeling the need to disregard social and legal rules. In particular, Asperger’s disorder individuals who have offended possess characteristics that exacerbate their risk of sexual offending in certain contexts: poor empathy, failure to develop peer appropriate relationships, deficits regarding stable emotional relationships, persistent preoccupation with parts/objects (Fabian, 2011; Haskins and Silva, 2006).
Name of Tool Assessment of Risk Manageability for Individuals with Developmental and Intellectual Limitations who Offend Sexually (ARMIDILO S)
•The ARMIDILO S is a structured risk and management guideline instrument which assesses the risk of sexual recidivism in individuals diagnosed with intellectual and developmental disabilities. It is the first effort to view persons with IDPSB within the context and environment in which individuals are located (Blasingame et al., 2014; Lindsay et al., 2018).
18+ Assessor Qualifications
Category Responsivity Issues (Awaiting Validation)
Year 2009 Description
•The tool is intended for males 18 and older who have engaged in sexual offending behaviour which may or may not have been adjudicated. It applies to individuals who have borderline intellectual functioning (i.e. IQ between 70 and 80 with adaptive functioning deficits) or are intellectually disabled (i.e. males with onset of cognitive impairment before the age of 18 reflected by an IQ score below 70 and have adaptive functioning deficits) There is currently no supporting evidence to suggest the ARMIDILO S can be applied to other offending populations: non ID, female, youth and forensic mental health (Boer et al., 2013).
•The ARMIDILO S only uses dynamic risk factors. The tool consists of 30 stable and acute items. The stable items reflect the persistent characteristics of the individual. The acute items represent rapidly changing contextual factors that signal the onset of offending behaviours. The stable and acute items are further divided into four subscales relating to ‘environmental’ and ‘client’ related factors: 1) stable dynamic environmental subscales (e.g. attitudes towards ID intellectuals, etc.); 2) acute dynamic environment subscales (e.g. access to intoxicants, etc.); 3) stable dynamic client subscale (e.g. compliance with treatment and supervision, etc.; 4) acute dynamic client subscale (e.g. victim access, etc.) (Boer et al., 2013).
RATED page updated: August 2019 © Risk Management Authority 2019
Author / Publisher Boer and Colleagues
•Each item is considered as both a risk and a protective factor. Items are scored on a 5 point scale from 2 for reducing risk through to +2 for an increase in risk. Once scored, the tool generates four risk ratings which include: (1) Actuarial Risk Rating (i.e. ratings obtained from other standardised actuarial tools such as the RRASOR), (2) Risk Rating, (3) Protective Rating and (4) Adjusted Risk Rating (i.e. consideration for other three ratings). Overall, risk manageability is defined as the ‘current dynamic risk manageability estimate,’ which is the ability of the individual to manage their dynamic factors adjusted by the actuarial risk baseline and the individual’s structured clinical risk estimate (Craig et al., 2008).
Age Appropriateness
General Notes
should have the relevant training and experience in administering and interpreting risk assessments in relation to individuals diagnosed with learning disabilities who are at risk of sexual violence.
Tool Development
•In an unpublished thesis by Sindall (2012), the ARMIDILO S was used in a sample of 16 individuals with intellectual disabilities who had committed sexual offences. The AUC was found to be 0.83, with the risk total (0.83), stable items (0.837), client items (0.86) and stable client items (0.85) all showing good predictive accuracy.
•A study by Lofthouse et al. (2013) administered the various risk assessment tools to sixty four adult males who had ID and a history of sexual offending in a six year follow up study. It was found that the ARMIDILO S yielded the best prediction of sexual reoffending with an AUC of .92 compared to other established risk assessment tools which included the Static 99 (AUC = .74) and the VRAG (AUC = .58). The authors surmised that predictive value of the tool may be attributable to it specifically being designed for individuals with ID, as well as its inclusion of dynamic variables.
•Blacker et al. (2011) assessed the predictive validity of the RRASOR, SVR 20, RM2000 V and ARMIDILO S on 88 individuals, half of which had committed sexual offences and had borderline levels of intellectual functioning with an IQ of 70 80. The ARMIDILO S was found to be the best predictor for offending in those with special needs, generating AUCs of .60 and .73 for the stable and acute scores of the instrument respectively. Having said that, this study did have missing information for the environmental variables, something which would have affected the validity of testing.
RATED page updated: August 2019 © Risk Management Authority 2019
•In their review of the literature, Pryboda and colleagues (2015) found that the ARMIDILO S showed superior predictive accuracy of the RM2000 when applied to those with intellectual disabilities. The authors suggest this could be due to the ARMIDILO S considering protective factors, meaning it can be used for short term risk management planning and long term risk predictions.
•The original tool was developed in 2004 and expanded in 2013 by Boer and colleagues to include a greater range of issues (e.g. victim availability and access, staff attitudes towards individuals with ID). The rationale for the tool was to create an ID specific instrument to meet the needs of these types of individuals. Moreover, it was felt that the inclusion of dynamic environmental and client variables would better inform the formulation of risk management plans for individuals (Boer et al., 2013).
•Lindsay et al. (2018) carried out a study applying the ARMIDILO S to four individuals with intellectual developmental disabilities. For two of the participants, restrictive placements were avoided because of the data generated on protective factors.
The ARMIDILO S is designed to assist support workers, case managers, guardians, home providers, clinicians and program administrators in the identification and management of risk (Boer et al., Assessors2013).
•The tool is conceptualised as part of a comprehensive assessment approach. It is, hence, recommended that the ARMIDILO S be used in conjunction with other actuarial and structured professional judgement measures. The authors advise that the appropriate caveats and caution is
•A doctoral thesis by Cookman (2010) found that the ARMIDILO S had significant correlations with the Stable 2007 and Acute 2009, suggesting concurrent validity is present.
•For more information, please visit the following website: www.armidilo.net
applied if using the instrument on an individual who has committed non sexual offences (Boer et al., 2013).
RATED page updated: August 2019 © Risk Management Authority 2019
•The ARMIDILO S is unique in that it examines both client and environmental dynamic variables (Lofthouse et al., 2013).
•A variation of the tool, the ARMIDILO G, has been developed to assess general recidivism in those with ID. The ARMIDILO G was found to have good predictive accuracy for a sample of 139 individuals with an intellectual disability and a history of offending in doctoral research by Frize (2015)
•Lindsay et al. (2018) recommend that protective factors are included in all risk assessments, maintaining that the protective scale can be a powerful support for the clinical case individuals with IDD who offend.
No specific age range.
Year 2004
RATED page updated: August 2019 © Risk Management Authority 2019
Name of Tool Dynamic Risk Assessment and Management System (DRAMS)
•Every item has been arranged along a continuum from no problem to serious problem. Furthermore, a ‘traffic light analogy’ replaces the traditional Likert scale with red for association with risk, amber for intermediate and green for least problematic (Gaskin, 2007).
Assessors should have the relevant training and experience in administering and interpreting risk assessments in relation to individuals diagnosed with learning disabilities who have offended.
Age Appropriateness
•The DRAMS can be scored by item, category and as a total score. The tool was intended to be used as part of a comprehensive approach to risk assessment and is best used idiomatically with individual clients (Lindsay et al., 2004).
Category Responsivity Issues (Awaiting Validation)
•The DRAMS is a 10 item risk assessment tool composed of proximal/dynamic risk factors for use with those diagnosed with learning disabilities.
Description
Author / Publisher Lindsay and Colleagues
Assessor Qualifications
•Although the measure was developed in relation to positive behavioural programmes, the measure can also be used with any therapeutic, educational, management or occupational regime, if deemed appropriate by clinicians (Lindsay et al., 2004).
•During periods where the DRAMS indicates low risk, normal day to day programmes in which the client engages can be implemented as usual. Conversely, if high dynamic risk is identified by the tool, such programmes can be suspended.
•The tool can be completed with the client.
•The DRAMS was developed by staff from The State Hospital in Scotland to be an instrument that could be easily understood and hence used by clients with IDs. Factors were developed based on literature from proximal/dynamic risk (Gaskin, 2007).
•It was originally designed to be used in conjunction with positive behavioural programmes implemented within secure settings. It assesses a number of proximal and dynamic risk variables, such as mood, self regulation, anti social behaviour and compliance with routine.
•It is used to assess the immediate risk of offending posed by the service user within secure settings. Given its dynamic nature, it should be completed at regular intervals.
Tool Development
RATED page updated: August 2019 © Risk Management Authority 2019
•Steptoe et al. (2008) the DRAMS had moderate inter rater reliability (rs = .46). The composite score achieved predictive accuracy in relation to violent incidents perpetrated in secure settings (AUC = .73).
General Notes
•It appears that DRAMS is effective across a range of settings (Lindsay et al., 2017).
It is noteworthy that studies have found differences in the items that are statistically significant predictors of future incidents. For instance, Lindsay et al. (2017) found that substance abuse and clinical items were significant predictors, likely due to the sample involving participants in the community where there is less regulations in place than in secure facilities.
•Lindsay et al. (2004) initial validation of the measure demonstrated high reliability for the composite score and moderate to high reliability for nine of the ten items presented in the measure. One item (‘therapeutic alliance’) retained poor reliability scores.
•Lindsay and colleagues (2017) tested the predictive accuracy of the DRAMS in a sample of 30 male participants. AUCs ranged from .52 .87; although the total scores score generated an AUC of .86.
Name of Tool Oxford Mental Illness and Violence Tool (OxMIV)
Assessor Qualifications
Category Responsivity (Awaiting Validation)
Description
Age Appropriateness
Author / Publisher Fazel, Wolf, Larsson, Lichtenstein, Mallett and Fanshawe
16+
RATED page updated: August 2019 © Risk Management Authority 2019
•The tool is designed to be used as an adjunct to clinical assessment. The prediction score can be used to identify those who are at low risk of violent offending. Individuals with a greater than 5% risk are categorised as ‘increased risk’ (Fazel et al., 2017).
•As the tool consists of mainly static factors, it should not be used to monitor within individual change in risk over time; instead it should be used as a cross sectional score at a particular time point (Fazel et al., 2017).
•The sixteen variables included in the tool: sex, age, previous drug abuse, previous alcohol abuse, previous self harm, highest education, parental drug or alcohol use, parental violent crime conviction (lifetime), sibling violent crime conviction (lifetime), current episode, recent antipsychotic treatment (within the last six months), recent antidepressant treatment (within the last six months), recent dependence treatment (within the last six months), personal income, benefit recipient (Fazel et al., 2017).
No specific training or qualifications are required to use the OxMIV. The determination of appropriate application and scoring to specific cases requires the judgement of a clinician (medical doctor, clinical psychologist or nurse). Since it relies on diagnostic and treatment information, it should not be administered by non clinical staff (Fazel et al., 2017).
•The variables considered for inclusion in the OxMIV were drawn from the existing body of evidence relating to criminal history and sociodemographic and clinical factors (Bonta, Blais and Wilson, 2014; Witt, van Dorn and Fazel, 2013).
•OxMIV predicts violent offending for individuals with schizophrenia spectrum and bipolar disorders within the next 12 months (Fazel et al., 2017).
Year 2017
Tool Development
•Using a sample of 75 158 Swedish individuals with a severe mental illness (schizophrenia spectrum or bipolar disorder, the OxMIV was developed to predict violent offending within 1 year of
•To make the tool less time consuming and complex, and also more reliable, OxMIV does not include information about risk factors that need to be collected from an interview, such as anger and victimisation issues and comorbid personality traits (Fazel et al., 2017).
•The tool is available in English, Greek, German, French and Chinese.
•The tool is freely available online: https://oxrisk.com/oxmiv/
General Notes
•It is not recommended for use in forensic psychiatric patients and released prisoners, as baseline risks and effect of risk factors could be different. The population in which the tool was tested was individuals with diagnosed with schizophrenia spectrum or bipolar disorder (Fazel et al., 2017).
RATED page updated: August 2019 © Risk Management Authority 2019
•OxMIV can be used at any point in a patient’s pathway. It cannot be used to detain individuals or extend their detention in the absence of other clinical factors and detailed assessment (Fazel et al., 2017).
•The sixteen items in the OxMIV are all of a retrospective nature and are not specifically designed for mapping an individual’s treatment process (Negatsch et al., 2019).
•OxMIV could be particularly useful for screening out low violent risk in general adult psychiatric services. The tool has a high negative predictive value (99.5%) in other words, of 200 individuals identified as low risk by the tool, 199 did not offend violently within in 1 year.
hospital discharge for inpatients or clinical contact with psychiatric services for outpatients. External validation carried out on 16 387 individuals showed good discrimination with a c index (equivalent to an AUC score) of 0.89. Calibration was also good (Brier score = 0.013). For risk of violent offending at 1 year, the sensitivity was 62% and the specificity was 94% (Fazel et al., 2017).
•Fazel et al. (2017) found that the strongest predictors of reoffending were convictions for previous violent crime, gender, and age. The weakest predictors were personal income and benefit receipt.
•Negatsch and colleagues (2019) applied the OxMIV to 474 male patients in a hospital in Germany with schizophrenia spectrum or bipolar disorder. These patients were classified into two groups: a violent group with 191 patients, where violence was defined as the aggressive behaviour of a patient that necessitated special observation; a non violent group with 283 patients. The OxMIV score was significantly higher in the violent group compared to the non violent one, showing the tool succeeded in predicting violent behaviour in male patients in imprisoned psychiatric settings. In particular, the items of ‘previous violent crime,’ ‘previous drug abuse’ and ‘previous alcohol abuse’ were all significant in predicting violent behaviour.
•The P SCAN does not provide a clinical diagnosis of psychopathy (see Warren, Chauhan and Murrie 2005).
Name of Tool P SCAN
RATED page updated: August 2019 © Risk Management Authority 2019
Category Responsivity Issues (Awaiting Validation)
•It is available for use by non clinicians within mental health and correctional settings.
General Notes
Author / Publisher Hare and Hervé
•Items are scored on a three point Likert scale according to the extent to which the client exhibits particular traits.
•Brackett, Jackson and Richards (2008) large correlation was observed between the composite scores for the PCL:R and the P SCAN (r = .49). Moderate to large coefficients obtained for the P SCAN subscales in relation to the rater’s experience with the construct of psychopathy.
Tool Development
•It is intended for use when it is not possible to conduct the full PCL:R or PCL:SV, taking around 10 to 15 minutes to complete. It does not, however, provide a clinical diagnosis
Year 1999
Age Appropriateness
•The P SCAN is a structured assessment that screens for psychopathic traits in individuals aged 18 and above.
18+ Assessor Qualifications
Description
•It is a 90 item checklist that explores three key facets of psychopathy; (1) interpersonal, (2) affective and (3) lifestyle.
It is designed for use by law enforcement, forensic and civil facilities, corrections, probations and any other places where determining the possible presence of psychopathic traits is of interest.
•Warren, Chauhan and Murrie (2005) the P SCAN obtained high internal consistency (∂ ≥ .96) in a sample of 115 females incarcerated in a maximum security prison. The inter rater reliability of the P SCAN varied: Total score was 0.40, Interpersonal Facet 0.51, Affect Facet 0.32 and Lifestyle Facert 0.24.
•Elwood, Poythress and Douglas (2004) found high internal consistency (∂ ≥ .90) for the three subscales in the P SCAN within a sample of university students.
•For more information, visit the following website: www.hare.org
RATED page updated: August 2019 © Risk Management Authority 2019
•Few validation studies on this tool at present.
Cooke, D. J. and Logan, C. (2018) ‘Capturing psychopathic personality: Penetrating the mask of sanity through clinical interview.’ In Patrick, C. J. (ed.) Handbook of Psychopathy. New York, Guilford, 189 210. Access Here
Florez, G., Ferrer, V., Garcia, L. S., Crespo, M. R., Perez, M., Saiz, P. A. and Cooke, D. J. (2018) ‘Clinician ratings of the Comprehensive Assessment of Psychopathic Personality (CAPP) in a representative sample of Spanish prison inmates: new validity evidence.’ PloS ONE 13(4), e0195483. Access Here.
D. J., Hart, S., Logan, C., & Michie, C. (2004). Comprehensive Assessment of Psychopathic Personality Institutional Rating Scale (CAPP IRS) (Unpublished manuscript). Glasgow, UK: Glasgow Caledonian University. [Not accessible]
De Page, L., Mercenier, S. and Titeca, P. (2018) ‘Assessing psychopathy in forensic schizophrenia spectrum disorders: validating the comprehensive assessment of the psychopathic personality institutional rating (CAPP IRS).’ Psychiatry Research 265, 303 308. Access Here
Cooke, D. J. and Sellbom, M. (2019) ‘An Examination of Psychopathy Checklist Revised Latent Factor Structure Via Exploratory Structural Equation Modeling.’ Psychological Assessment 31 (5), 581 591. Access Here.
Cooke D. J. and Logan C. (2015) ‘Capturing clinical complexity: Towards a personality oriented measure of psychopathy.’ Journal of Criminal Justice 43(4), 262 73. Access Here.
Cooke D. J., Hart, S. D., Logan, C. and Michie, C. (2012) ‘Explicating the Construct of Psychopathy: Development and Validation of a Conceptual Model, the Comprehensive Assessment of Psychopathic Personality (CAPP).’ International Journal of Forensic Mental Health 11(4), 242 52. Access Here
Kreis, M. K. F. (2009) Psychopathy in women: a multi method exploration of the construct using the Comprehensive Assessment of Psychopathic Personality (CAPP). Doctoral thesis. Glasgow, Scotland: Glasgow Caledonian University. Access Here
Cooke, D. J., Hart, S. D., Logan, C. and Michie, C. (under review) ‘Evaluating the Test Validity of the Comprehensive Assessment of Psychopathic Personality Symptoms Rating Scale.’ Psychological Assessment. [Not accessible]
Cooke,CAPP
Dawson, S., McCuish, E., Hart, S. D. and Corrado, R. R. (2012) ‘Critical Issues in the Assessment of Adolescent psychopathy: An Illustration Using Two Case Studies.’ International Journal of Forensic Mental Health 11(2), 63 79. Access Here
Kreis, M. K. F. and Cooke, D. J. (2011) ‘Capturing the psychopathic female: a prototypicality analysis of the comprehensive assessment of psychopathic personality (CAPP) across gender.’ Behavioral Sciences & the Law 29(5), 634 648. Access Here
VALIDATED TOOLS
Cooke, D. J. (2018) ‘Psychopathic Personality Disorder: Capturing an Elusive Concept.’ European Journal of Analytic Philosophy 14(1), 15 32. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Kreis, M. K. F., Cooke, D. J., Michie, C., Hoff, H. A. and Logan, C. (2012) ‘The Comprehensive Assessment of Psychopathic Personality (CAPP): Content Validation Using Prototypical Analysis.’ Journal of Personality Disorder 26(3), 402 413. Access Here.
Sellbom, M., Cooke, D. J. and Hart, S. D. (2015) ‘Construct Validity of the Comprehensive Assessment of Psychopathic Personality (CAPP) Concept Map: Getting Closer to the Core of Psychopathy.’ International Journal of Forensic Mental Health 14(3), 172 180. Access Here
Skeem, J. L. and Cooke, D. J. (2010) ‘Is criminal behavior a central component of psychopathy? Conceptual directions for resolving the debate.’ Psychological Assessment 22(2), 433 445. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Sandvik, A. M., Hansen, A. L., Kristensen, M. V., Johnsen, B. H., Logan, C. and Thornton, D. (2012) ‘Assessment of Psychopathy: Inter correlations between Psychopathy Checklist Revised, Comprehensive Assessment of Psychopathic Personality Institutional Rating Scale, and Self Report of Psychopathy Scale III.’ International Journal of Forensic Mental Health 11(4), 280 288. Access Here
McCuish, E. C., Hanniball, K. B. and Corrado, R. (2019) ‘The Assessment of Psychopathic Personality Disturbance Among Adolescent Male Offenders.’ International Journal of Forensic Mental Health 18(1), 35 49. Access Here
Nikolovia, N. L. (2009) The Psychopathic Personality Inventory Revised: Evaluation of Its Psychometric Properties, Incremental Validity, and Moderating Effects of Gender in a Correctional Sample. Doctoral thesis. British Columbia, Canada: Simon Fraser University. Access Here.
Pauli, M., Essemyr, K., Sorman, K., Howner, K., Gustavsson, P. and Liljeberg, J. (2018) ‘Gendered expressions of psychopathy: correctional staff’s perceptions of the CAPP and CABP models.’ International Journal of Forensic Mental Health 17(2), 97 110. Access Here
Pedersen, L., Kunz, C., Elsass, P. and Rasmussen, K. (2010) ‘Psychopathy as a risk factor for violent recidivism investigating the Psychopathy Checklist Screening Version (PCL:SV) and the Comprehensive Assessment of Psychopathic Personality (CAPP) in a forensic psychiatric setting.’ International Journal of Forensic Mental Health 9(4), 308 315. Access Here
Sea, J. (2018) Cross cultural Generalizability of the Comprehensive Assessment of Psychopathic Personality (CAPP) in South Korea. Doctoral thesis. Simon Fraser University: British Columbia, Canada. Access Here.
McCormick, A. (2004) Interrater reliability of the Comprehensive Assessment of Psychopathic Personality Disorder Among a Sample of Incarcerated Serious and Violent Young Offenders. Master of Arts dissertation. British Columbia, Canada: Simon Fraser University. Access Here
Kreis, M. K. F. and Cooke, D. J. (2012) ‘The Manifestation of Psychopathic Traits in Women: An Exploration Using Case Examples.’ International Journal of Forensic Mental Health 11(4), 267 279. Access Here
McCuish, E., Bouchard, M., Beauregard, E. and Corrado, R. (2019) ‘A Network Approach to Understanding the Structure of Core Symptoms. Of Psychopathic Personality Disturbance in Adolescents Offenders.’ Journal of Abnormal Child Psychology 47(9), 1467 1482. Access Here.
Loranger, A. W., Sartorius, N., Andreoli, A., Berger, P., Buchheim, P., Channabasavanna, S. M., Coid, B. D., Dahl, A., Diekstra, R. F. W., Ferguson, B., Jacobsberg, L. B., Mombour, W., Pull, C., Ono, Y. and Regier, D. A. (1994) ‘The International Personality Disorder Examination: The World Health Organization/Alcohol, Drug Abuse, and Mental Health Administration international pilot study of personality disorders.’ Archives of General Psychiatry 51(3), 215 224. Access Here.
IPDE
Loranger A. W., Janca, A. and Sartorius, N. (1997) Assessment and Diagnosis Of Personality Disorders: The ICD I0 International Personality Disorder Examination (IPDE). Cambridge: Cambridge University Press. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Haider, I., Ijaz, F. B., Fareeha, H., Ayub, M., Irfan, M. and Naeem, F. (2014) ‘Reliability of the ICD 10 international personality disorder examination (Urdu translation): A Preliminary Study.’ Pakistan Journal of Medical Sciences 30(6), 1372 1376. Access Here
Cooke, D. J. and Hart, S. D. (2004). Personality disorders. In E. C. Johnstone, S. M. Lawrie, D. C. Owens and M. Sharpe (eds.), Companion to psychiatric studies (7th edition). Edinburgh, UK: Elsevier, 502 526. Access Here.
Janca, A. and Pull, C. (1997) ‘Description of centres participating in the IPDE field trial.’ In Loranger, A. W., Janca, A., and Sartorius, N. Assessment and diagnosis of personality disorders: the ICD 10 international personality disorder examination (IPDE). New York: Cambridge University Press, 61 69. Access Here
Magallón Neri E. M., Forns, M., Canalda, G., De La Fuente, J. E., García, R., González, E., Lara, A., Castro Fornieles, J. (2013) ‘Usefulness of the International Personality Disorder Examination Screening Questionnaire for borderline and impulsive personality pathology in adolescents.’ Comprehensive Psychiatry 54(3), 301 308. Access Here.
Mann, A. H., Raven, P., Pilgrim, J., Khanna, S., Velayudham, A., Suresh, K. P., Channabasavanna, S.M., Janca, A. and Sartorius, N. (1999) ‘An assessment of the Standardized Assessment of Personality as a screening instrument for the International Personality Disorder Examination: a comparison of informant and patient assessment for personality disorder.’ Psychological Medicine 29(4), 985 989. Access Here
El Rufaie, E. F., Al Sabosy, M., Abuzeid, M. S. and Ghubash, R. (2002) ‘Personality profile among primary care patients: experimenting with the Arabic IPDE ICD 10.’ Acta Psychiatrica Scandinavica 105(1), 37 41. Access Here
Fountoulakis, K. N., lacovides, A., loaannidou, C., Bascialla, F., Nimatoudis, I., Kaprinis, G., Janca, A. and Dahl, A. (2002) ‘Reliability and cultural applicability of the Greek version of the International Personality Disorders Examination.’ BMC Psychiatry 2(6). Access Here.
Clarkin, J. F., Levy, K. N., Lenzenweger, M. F. and Kernberg, O. F. (2007) ‘Evaluating Three Treatments for Borderline Personality Disorder: A Multiwave Study.’ The American Journal of Psychiatry 164(6), 922 928. Access Here
Kvale, S. and Brinkmann, S. (2009) InterViews: Learning the Craft of Qualitative Research. Los Angeles: Sage. Access Here
Coid, J., Yang, M., Ullrich, S., Zhang, T., Roberts, A., Roberts, C., Rogers, R. and Farrington, D. (2007) Predicting and understanding risk of re offending: The prisoner cohort study. Research Summary 6. London, UK: Ministry of Justice. Access Here.
Berrios, G. E. (1996) The History of Mental Symptoms: Descriptive Psychopathology Since the Nineteenth Century. Cambridge: Cambridge University Press. Access Here.
Blais, J., Forth, A. E. and Hare, R. S. (2017) ‘Examining the inter rater reliability of the Hare Psychopathy Checklist revised across a large sample of trained raters.’ Psychological Assessment 29(6), 762 775. Access Here
Cooke, D. J., Michie, C., Hart, S. D. and Clark, D. (2005) ‘Assessing psychopathy in the UK: concerns about cross cultural generalisability.’ British Journal of Psychiatry 186(4), 335 341. Access Here
DeMatteo, D. and Edens, J. F. (2006) ‘The Role and Relevance of the Psychopathy Checklist Revised in Court.’ Psychology, Public Policy and Law 12(2), 214 241. Access Here
Sharan P., Kulhara P., Verma S. K. and Mohanty, M. (2002) ‘Reliability of the ICD 10 International Personality Disorder Examination (IPDE) (Hindi Version): A preliminary study.’ Indian Journal of Psychiatry 44(4), 362 364. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Clecklley, H. (1976) The Mask of Sanity: An Attempt to Clarify Some Issues about the So called Psychopathic Personality (5th edition). St Louis, MO: Mosby. Access Here
Coid, J., Yang, M., Ullrich, S., Roberts, A., Moran, P., Bebbington, P., Brugha, T., Jenkins, R., Farrell, M., Lewis, G., Singleton, N. and Hare, R. (2009) ‘Psychopathy among prisoners in England and Wales.’ International Journal of Law and Psychiatry 32(3), 134 141. Access Here
Cooke, D. J., Michie, C. and Ryan, J. (2001) Evaluating Risk for Violence: A Preliminary Study of the HCR 20, PCL R and VRAG in a Scottish Prison Sample, Edinburgh: Scottish Prison Service Occasional Paper 5/2001. [Not accessible]
PCL R Abbiati, M., Palix, J., Gasser, J. and Moulin, V. (2018) ‘Predicting psychically violent misconduct in prison: A comparison of four risk assessment instruments.’ Behavioral Sciences & the Law 37(1), 1 77. Access Here
Cox, J., DeMatteo, D. S. and Foster, E. E. (2010) ‘The effect of the Psychopathy Checklist Revised in capital cases: mock jurors’ responses to the label of psychopathy.’ Behavioural Sciences & the Law 28(6), 878 891. Access Here.
Slade, K. and Forrester, A. (2013) ‘Measuring IPDE SQ personality disorder prevalence in pre sentence and early stage prison populations, with subtype estimated.’ International Journal of Law and Psychiatry 36(3 4), 207 212. Access Here.
Daffern, M. (2007) ‘The predictive validity and practical utility of structured schemes used to assess risk for aggression in psychiatric inpatient settings.’ Aggression and Violent Behavior 12(1), 116 130. Access Here
Cooke, D. J. and Michie, C. (1999). ‘Psychopathy across cultures: North America and Scotland compared.’ Journal of Abnormal Psychology 108(1), 55 68. Access Here
Gardner, B. O., Boccaccini, M. T. and Murrie, D. C. (2018) ‘Which PCL R scores best predict forensic clinicians’ opinions of offender risk?’ Criminal Justice and Behavior 45(9), 1404 1419. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
González, Moraga, F. R., Garcia, D., Billstedt, E. and Wallinius, M. (2019) ‘Facets of Psychopathy, Intelligence and Aggressive Antisocial Behaviors in Young Violent Offenders.’ Frontiers in Psychology 10, 984. Access Here.
Hawes, S. W., Boccaccini, M. T. and Murrie, D. C. (2013) ‘Psychopathy and the combination of psychopathy and sexual deviance as predictors of sexual recidivism: Meta analytic findings
Gray, N.S. and R. J. Snowden. (2016) ‘Psychopathy in women: prediction of criminality and violence in UK and USA psychiatric patients resident in the community.’ Psychiatry Research 237, 339 343. Access Here
Douglas, K. S. and Reeves, K. A. (2010) ‘Historical Clinical Risk Management 20 (Hcr 20) Violence Risk Assessment Scheme: Rationale, Application and Empirical Overview.’ in R. K. Otto and K. S. Douglas. Handbook of Violence Risk Assessment. New York, London: Routledge, 147 186. Access Here
Florez, G., Ferrer, V., Garcia, L. G., Crespo, M. R., Perez, M., Saiz, P. A. and Cooke, D. J. (2018) ‘Novel validity evidence of the Psychopathy Checklist Revised (PCL R) in a representative sample of 204 Spanish inmates.’ Forensic Science International 291, 75 83. Access Here.
Hare, R. D. (2003) The Hare Psychopathy Checklist Revised (2nd edition). North Tonawanda, New York: Multi Health Systems. [Not accessible]
de Vogel, V., Bruggeman, M. and Lancel, M. (2019) ‘Gender sensitive Violence Risk Assessment: Predictive Validity of Six Tools in Female Forensic Psychiatric Patients.’ Criminal Justice and Behavior. Access Here
DeMatteo, D., Edens, J. F., Gallaway, M., Cox, J., Smith, S. T. and D. Forman. (2014) ‘The role and reliability of the Psychopathy Checklist Revised in the U.S. sexually violent predator evaluations: a case law survey.’ Law and Human Behaviour 38(3), 248 255. Access Here.
Farrington, D. P., Jolliffe, D. and Johnstone, L. (2008) Assessing Violence Risk: A Framework for Practice (Final Report). Paisley, UK: Risk Management Authority. Access Here
Hare, R. D. (1998) ‘The Hare PCL R: Some issues concerning its use and misuse.’ Legal and Criminological Psychology 3(1), 99 119. Access Here.
DeMatteo, D., Edens, J. F. and Hart, A. (2010). The use of measures of psychopathy in violence risk assessment. In R. K. Otto and K. S. Douglas. Handbook of Violence Risk Assessment. New York, London: Routledge, 19 40. Access Here
Hare, R. D. (1991) The Hare Psychopathy Checklist Revised. North Tonawanda, New York: Multi Health Systems. [Not accessible]
Hare, R. D. and Neumann, C. S. (2009) ‘Psychopathy: Assessment and Forensic Implications.’ Canadian Journal of Psychiatry 54(12), 791 802. Access Here.
Ismail, G. and Looman, J. (2016) ‘Field inter rater reliability of the psychopathy checklist revised.’ International Journal of Offender Therapy and Comparative Criminology 62(2), 468 481. Access Here
Olver, M. E., Neumann, C. S., Wong, S. C. and Hare, R. D. (2013) ‘The Structural and predictive properties of the psychopathy checklist revised in Canadian Aboriginal and non Aboriginal offenders.’ Psychological Assessment 25(1), 167 179. Access Here.
Olver, M. E., Neumann, C. S., Sewall, L. A., Lewis, K., Hare, R. D. and Wong, S. C. P. (2018) ‘A comprehensive examination of the psychometric properties of the Hare Psychopathy Checklist Revised in a Canadian multisite sample of indigenous and non indigenous offenders.’ Psychological Assessment 30(6), 779 792. Access Here.
Lloyd, C. D., Clark, H. J., and Forth, A. E. (2010) ‘Psychopathy, expert testimony and indeterminate sentences: Exploring the relationship between psychopathy checklist revised testimony and trial outcome in Canada.’ Legal and Criminological Psychology 15(2), 323 339. Access Here
Morrissey, C., Mooney, P., Hogue, T. E., Lindsay, W. R. and Taylor, J. L. (2007) ‘Predictive validity of the PCL R for offenders with intellectual disability in a high security hospital: Treatment progress.’ Journal of Intellectual & Developmental Disability 32(2), 125 133. Access Here.
Laurell, J. and Dåderman, A. M. (2007) ‘Psychopathy (PCL R) in a forensic psychiatric sample of homicide offenders: Some reliability issues.’ International Journal of Law and Psychiatry, 30(2), 127 135. Access Here.
Morrissey, C., Cooke, D., Michie, C., Hollin, C., Hogue, T., Lindsay, W. R. and Taylor, J. L. (2010) ‘Structural, Item, and Test Generalizability of the Psychopathy Checklist Revised to Offenders With Intellectual Disabilities.’ Assessment 17(1), 16 29. Access Here
Olver, M. E. and Wong, S. C. (2006) ‘Psychopathy, Sexual Deviance, and Recidivism Among Sex Offenders.’ Sexual Abuse 18(1), 65 82. Access Here
using the Psychopathy Checklist Revised.’ Psychological Assessment 25(1), 233 243. Access Here.
Krstic, S., Neumann, C. S., Roy, S., Robertson, C. A., Knight, R. A. and R. D. Hare. (2017) ‘Using latent variable and person centred approaches to examine the role of psychopathic traits in sex offenders.’ Personality Disorders: Theory, Research, and Treatment 9(3), 207 216. Access Here
Logan, C. and Blackburn, R. (2009) ‘Mental disorder in violent women in secure settings: Potential relevance to risk for future violence.’ International Journal of Law and Psychiatry 32(1), 31 38. Access Here.
McDermott, B. E., Edens, J. F., Quanbeck, C. D., Busse, D. and Scott, C. L. (2008) ‘Examining the Role of Static and Dynamic Risk Factors in the Prediction of Inpatient Violence: Variable and Person Focused Analyses.’ Law and Human Behavior 32(4), 325 338. Access Here
NHS England and National Offender Management Service. (2015) Working with offenders with personality disorder: a practitioner’s guide (2nd edition). Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Tsang, S., Piquero, A. R. and E. Caufman. (2014) ‘An examination of the psychopathy checklist: youth version (PCL: YV) among male adolescent offenders: an item response theory analysis.’ Psychological Assessment 36(4), 1333 1346. Access Here.
Campbell, M. A., French, S., and Gendreau, P. (2009) ‘The Prediction of Violence in Adult Offenders: A Meta Analytic Comparison of Instruments and Methods of Assessment.’ Criminal Justice and Behavior 36(6), 567 590. Access Here.
Sullivan, E. A., Abramowitz, C. S., Lopez, M. and Kosson, D. S. (2006) ‘Reliability and construct validity of the psychopathy checklist revised for Latino, European American, and African American male inmates.’ Psychological Assessment 18(4), 382 392. Access Here
ArbachPCL:SV Lucioni, K., Andrés Pueyo, A., Pomarol Clotet, E. and Gomar Soñes, J. (2011) ‘Predicting violence in psychiatric inpatients: a prospective study with the HCR 20 violence risk assessment scheme.’ The Journal of Forensic Psychiatry & Psychology 22(2), 203 222. Access Here.
Rettenberger, M., Matthes, A., Boer, D. P. and Eher, R. (2010) ‘Prospective Actuarial Risk Assessment: A Comparison of Five Risk Assessment Instruments in Different Sexual Offender Subtypes.’ International Journal of Offender Therapy and Comparative Criminology 54(2), 169 186. Access Here.
Pichot, P. (1978) ‘Psychopathic Behaviour: a historical overview. In Hare, R. D. and Schalling, D.(eds.) Psychopathic Behaviour: Approaches to Research, 55 70. Chichester, UK: John Wiley & Sons, Inc. [Not accessible]
Skeem, J. L., Edens, J. F., Camp, J. and Colwell, L. H. (2004) ‘Are there Ethnic Differences in Levels of Psychopathy? A Meta Analysis.’ Law and Human Behavior 28(5), 505 527. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Brazil K. J. and Forth A. E. (2016) ‘Hare Psychopathy Checklist.’ In Zeigler Hill V. and Shackelford T.(eds.) Encyclopaedia of Personality and Individual Differences. New York, NY: Springer, 1 4. Access Here
Schaap, G., Lammers, S. and de Vogel, V. (2009) ‘Risk assessment in female forensic psychiatric patients: a quasi prospective study into the validity of the HCR 20 and PCL R.’ The Journal of Forensic Psychiatry & Psychology 20(3), 354 365. Access Here
Weizmann Henelius, G., Putkonen, H., Gronrous, M., Lindberg, N., Eronen, M. and Hakkanen Nyholm, H. (2010) ‘Examination of psychopathy in female homicide offenders confirmatory analyses of the PCL R.’ International Journal of Law & Psychiatry 33, 177 183. Access Here.
Vitale, J. E., Smith, S. S., Brinkley, C. A. and Newman, J. P. (2002) ‘The reliability and validity of the Psychopathy Checklist Revised in a sample of female offenders.’ Criminal Justice and Behavior 29(2), 202 231. Access Here
Cooke, D. J., Michie, C., Hart, S. D. and Hare, R. D. (1999) ‘Evaluating the Screening Version of the Hare Psychopathy Checklist Revised Screening Version (PCL:SV): An Item Response Theory Analysis.’ Psychological Assessment 11(1), 3 13. Access Here
Higgs, T., Tully, R. J. and Browne, K. D. (2018) ‘Psychometric Properties in Forensic Application of the Screening Version of the Psychopathy Checklist.’ International Journal of Offender Therapy and Comparative Criminology 62(7), 1869 1887. Access Here.
Ho, H., Thomson, L. and Darjee, R. (2009) ‘Violence risk assessment: the use of the PCL SV, HCR 20, and VRAG to predict violence in mentally disordered offenders discharged from a medium secure unit in Scotland.’ The Journal of Forensic Psychiatry & Psychology 20(4), 523 541. Access Here
Hart, S. D., Cox, D. N. and Hare, R. D. (1995) The Psychopathy Checklist: Screening Version. Toronto, ON: Multi Health Systems. [Not accessible]
Cullen, A. E., Soria, C., Clarke, A. Y., Dean, K. and Fahy, T. (2011) ‘Factors Predicting Dropout From the Reasoning and Rehabilitation Program With Mentally Disordered Offenders.’ Criminal Justice and Behavior 38(3), 217 230. Access Here.
Dietiker, J., Dittmann, V. and Graf, M. (2007) ‘Gutachterliche risikoeinschätzung bei sexualstraftätern. Anwendbarkeit von PCL SV, HCR 20+3 und SVR 20 [Risk assessment of sex offenders in a German speaking sample. Applicability of PCL SV, HCR 20+3, and SVR 20].’ Der Nervenarzt 78(1), 53 61. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
de Oliveira Souza, R., Hare, R. D., Bramati, I. E., Garrido, G. J., Azevedo Ignácio, F., Tovar Moll, F.and Moll, J. (2008) ‘Psychopathy as a disorder of the moral brain: fronto temporo limbic grey matter reductions demonstrated by voxel based morphometry.’ NeuroImage 40(3), 1202 1213. Access Here
Doyle, M., Carter, S., Shaw, J. and Dolan, M. (2012) ‘Predicting community violence from patients discharged from acute mental health units in England.’ Social Psychiatry and Psychiatric Epidemiology 47(4), 627 637. Access Here
Gray, N. S., Snowden, R. J., MacCulloch, S., Phillips, H., Taylor, J. and MacCulloch, M. J. (2004) ‘Relative Efficacy of Criminological, Clinical, and Personality Measures of Future Risk of Offending in Mentally Disordered Offenders: A Comparative Study of HCR 20, PCL:SV, and OGRS.’ Journal of Consulting and Clinical Psychology 72(3), 523 530. Access Here
Gray, N., Fitzgerald, S., Taylor, J., Macculloch, M. and Snowden, R. (2007) ‘Predicting Future Reconviction in Offenders With Intellectual Disabilities: The Predictive Efficacy of VRAG, PCL SV, and the HCR 20.’ Psychological Assessment. 19(4), 474 479. Access Here.
Douglas, K. S., Strand, S., Belfrage, H., Fransson, G. and Levander, S. (2005) ‘Reliability and Validity Evaluation of the Psychopathy Checklist: Screening Version (PCL:SV) in Swedish correctional and forensic psychiatric samples.’ Assessment 12(2), 145 161. Access Here.
Guy, L. S. and Douglas, K. S. (2006) ‘Examining the utility of the PCL:SV as a screening measure using competing factor models of psychopathy.’ Psychological Assessment 18(2), 225 230. Access Here
Cox, J., DeMatteo, D. S. and Foster, E. E. (2010) ‘The effect of the Psychopathy Checklist Revised in capital cases: mock jurors’ responses to the label of psychopathy.’ Behavioural Sciences & the Law 28(6), 878 891. Access Here
Lloyd, C. D., Clark, H. J., and Forth, A. E. (2010) ‘Psychopathy, expert testimony and indeterminate sentences: Exploring the relationship between psychopathy checklist revised testimony and trial outcome in Canada.’ Legal and Criminological Psychology 15(2), 323 339.
Brazil, K. and Forth, A. (2016) ‘Psychopathy Checklist: Youth Version (PCL:YV).’ In V.Zeigler Hill and T.K. Shackelford (eds.), Encyclopedia of Personality and Individual Differences. New York, NY: Springer, 1 4. Access Here.
Žukauskienė, R., Laurinavičius, A. and Čėsnienė, I. (2010) ‘Testing Factorial Structure and Validity of the PCL:SV in Lithuanian Prison Population.’ Journal of Psychopathology and Behavioral Assessment 32(3), 363 372. Access Here
Dawson, S., McCuish, E., Hart, S. D. and Corrado, R. R. (2012) ‘Critical Issues in the Assessment of Adolescent psychopathy: An Illustration Using Two Case Studies.’ International Journal of Forensic Mental Health 11(2), 63 79. Access Here.
Catchpole, R. E. H. and Gretton, H. M. (2003) ‘The predictive validity of risk assessment with violent young offenders. A 1 year examination of criminal outcome.’ Criminal Justice and Behavior 30(6), 688 708. Access Here
Cooke, D., Michie, C., Hollin, C., Hogue, T., Lindsay, W. R. and Taylor, J. L. (2010) ‘Structural, Item, and Test Generalizability of the Psychopathy Checklist Revised to Offenders With Intellectual Disabilities.’ Assessment 17(1), 16 29. Access Here
Corrado, R. R., Vincent, G. M., Hart, S. D. and Cohen, I. M. (2004) ‘Predictive validity of the Psychopathy checklist: Youth version for general and violent recidivism.’ Behavioral Sciences and the Law, 22(1), 5 22. Access Here
Stoll, C. B., Boillat, C., Pflueger, M. O., Graf, M. and Rosburg, T. (2019) ‘Psychopathy, Neuroticism and Abusive Behavior in Low Risk Child Sex Offenders.’ Journal of Child Sexual Abuse. Access Here
Bauer, D. L. A., Whitman, L. and Kosson D. S. (2011) ‘Reliability and construct validity of Psychopathy Checklist: Youth Version scores in incarcerated adolescent girls.’ Criminal Justice and Behavior 38(10), 965 987. Access Here
Access Morrissey,Here.C.,
PCL:YV
Richards, H. J., Casey, J. O. and Lucente, S. W. (2003) ‘Psychopathy And Treatment Response In Incarcerated Female Substance Abusers.’ Criminal Justice and Behavior 30(2), 251 276. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Cauffman, E., Kimonis, E. R., Dmitrieva, J. and Monahan, K. C. (2009) ‘A multimethod assessment of juvenile psychopathy: Comparing the predictive utility of the PCL:YV, YPI, and NEO PRI.’ Psychological Assessment 21(4), 528 542. Access Here.
Dolan, M. C. and Rennie, C. E. (2006) ‘Reliability and validity of the psychopathy checklist: Youth version in a UK sample of conduct disordered boys.’ Personality and Individual Differences 41(4), 779 789. Access Here
Howard, P. (2007) OASys General Reoffending Predictor and OASys Violent/Sexual Predictor Unpublished manuscript. [Not accessible]
Pechorro, P., Barroso, R., Maroco, J., Vieira, R. X. and Goncalves, R. A. (2015) ‘Psychometric properties of the Psychopathy Checklist: Youth Version Among Portuguese Juvenile Delinquents.’ International Journal of Offender therapy and Comparative Criminology 59(12), 1322 1337. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Marsh, A. A., Finger, E. C., Schechter, J. C., Jurkowitz, I. T. N., Reid, M. E. and Blair, R. J. R. (2011) ‘Adolescents with psychopathic traits report reductions in physiological responses to fear.’ The Journal of Child Psychology and Psychiatry 52(8), 834 841. Access Here.
Marshall, J., Egan, V., English, M. and Jones, R. M. (2006) ‘The relative validity of psychopathy versus risk/needs based assessments in the prediction of adolescent offending behaviour.’ Legal and Criminological Psychology 11(2), 197 210. Access Here
McCuish, E. C., Hanniball, K. B. and Corrado, R. (2019) ‘The Assessment of Psychopathic Personality Disturbance Among Adolescent Male Offenders.’ International Journal of Forensic Mental Health 18(1), 35 49. Access Here
Edens, J. F., Petrila, J. and Buffington Vollum, J. K. (2001) ‘Psychopathy and the Death Penalty: Can the Psychopathy Checklist Revised Identify Offenders Who Represent “A Continuing Threat to Society”?’ The Journal of Psychiatry & Law 29(4), 433 481. Access Here.
Hemphälä, M., Kosson, D. S., Westerman, J. and Hodgins, S. (2015) ‘Stability and predictors of psychopathic traits from mid adolescence through early adulthood among men and women treated for substance misuse in adolescence.’ Scandinavian Journal of Psychology 56(6), 649 658. Access Here
Edens, J. F. and Cahill, M. A. (2007) ‘Psychopathy in Adolescence and Criminal Recidivism in Young Adulthood: Longitudinal Results From a Multiethnic Sample of Youthful Offenders.’ Assessment 14(1), 57 64. Access Here
Edens, J. F. (2001) ‘Misuses of the Hare Psychopathy Checklist Revised in court: Two case examples.’ Journal of Interpersonal Violence 16(10), 1082 1093. Access Here
McCuish, E. C., Matthewsius, J. R., Lussier, P. and Corrado, R. R. (2018) ‘The cross cultural generalizability of the Psychopathy Checklist: Youth Version for adjudicated indigenous youth.’ Psychological Assessment 30(2), 192 203. Access Here.
Dolan, M. C. and Rennie, C. E. (2008) ‘The Structured Assessment of Violence Risk in Youth as a predictor of recidivism in a United Kingdom cohort of adolescent offenders with conduct disorder.’ Psychological Assessment 20(1), 35 46. Access Here
Forth, A. E., Kosson, D. S. and Hare, R. D. (2003) The Hare Psychopathy Checklist: Youth Version. Toronto: Multi Health Systems. [Not accessible]
Douglas, K. S., Epstein, M. E. and Poythress, N. G. (2008) ‘Criminal recidivism among juvenile offenders: Testing the incremental and predictive validity of three measures of psychopathic features.’ Law and Human Behavior 32(5), 423 438. Access Here.
Kotler, J. and McMahon, R. J. (2010) ‘Assessment of child and adolescent psychopathy.’ In R. T.Salekin and D. R. Lynam (eds.), Handbook of Child and Adolescent Psychopathy. New York, NY: Guilford Press, 79 110. Access Here
Stoesz, B. M., Montgomery, J. M., Smont, S. L. and Hellsten, L. A. M. (2011) ‘Review of five instruments for the assessment of Asperger’s disorder in adults.’ The Clinical Neuropsychologist 25(3), 376 401. Access Here
RATED page updated: August 2019 © Risk Management Authority 2019
Welsh J., Schmidt F., McKinnon L., Chatta H. K. and Meyers J. R. (2008) ‘A comparative study of adolescent risk assessment instruments: Predictive and incremental validity.’ Assessment 15(1), 104 115. Access Here.
Barry Walsh J. B. and Mullen P. E. (2004) ‘Forensic aspects of Asperger’s syndrome.’ Journal of Forensic Psychiatry and Psychology 15(1), 96 107. Access Here.
Schmidt, F., McKinnon, L., Chattha, H. K. and Brownlee, K. (2006) ‘Concurrent and predictive validity of the Psychopathy Checklist: Youth version across gender and ethnicity.’ Psychological Assessment 18(4), 393 401. Access Here
Murphy, D. (2007) ‘Hare Psychopathy Checklist Revised profiles of male patients with Asperger’s Syndrome detained in high security psychiatric care.’ The Journal of Forensic Psychiatry and Psychology 18(1), 120 126. Access here.
Attwood,AAA
Shepherd, S. M., Luebbers, S., Ogloff, J. R. P., Fullam, R. and Dolan, M. (2014) ‘The Predictive Validity of Risk Assessment Approaches for Young Australian Offenders.’ Psychiatry, Psychology and Law 21 (5), 801 817. Access Here.
ARMIDILO S
Stockdale, K. C., Olver, M. E. and Wong, S. C. P. (2010) ‘The Psychopathy Checklist: Youth Version and adolescent and adult recidivism: Considerations with respect to gender, ethnicity, and age.’ Psychological Assessment 22(4), 768 781. Access Here
Haskins, B. G. and Silva, J. A. (2006) ‘Asperger’s Disorder and Criminal Behavior: Forensic Psychiatric Considerations.’ Journal of the American Academy of Psychiatry and the Law 34(3), 374 384. Access Here.
TOOLS AWAITING VALIDATION
Kenny, H. and Stansfield, A. J. (2016) ‘How useful are the Adult Asperger Assessment and AQ 10 within an adult clinical population of all intellectual abilities.’ Advances in Autism 2(3), 118 130. Access Here
T. (2007) The complete guide to Asperger's syndrome. London, England: Jessica Kingsley Publishers. Access Here
Fabian, J. (2011) ‘Assessing the Sex Offender With Asperger’s Disorder: A Forensic Psychological and Neuropsychological Perspective.’ Sex Offender Law Report 12(5), 65 80. [Not accessible]
Baron Cohen, S., Wheelwright, S., Robinson, J. and Woodbury Smith, M. (2005) ‘The Adult Asperger Assessment (AAA): A Diagnostic Method.’ Journal of Autism and Developmental Disorders 35(6), 807 819. Access Here.
Lindsay, W. R., Finlay, C., Steptoe, L., Haut, F. and Brewster, E. (2017) ‘Predictive validity of the dynamic risk assessment and management system in individuals with intellectual disability residing in the community.’ Psychology, Crime & Law 24(4), 391 399. Access Here
Boer, D. P., Haaven, J. L., Lambrick, F., Lindsay, W. R., McVilly, K. R., Saksalan, J. and Frize, M. (2013) The Assessment of Risk Manageability of Individuals with Developmental and Intellectual Limitations who Offend Sexually. Access Here
O. (2012) An exploratory validation study of a risk assessment tool for male sex offenders with an intellectual disability. Doctoral thesis. Kent, UK: Canterbury Christ Church University. Access Here
Gaskin,DRAMS
Blacker, J., Beech, A. R., Wilcox, D. T. and Boer, D. P. (2011) ‘The assessment of dynamic risk and recidivism in a sample of special needs sexual offenders.’ Psychology, Crime & Law 17(1), 75 92. Access Here
Lindsay, W. R., Murphy, L., Smith, G., Murphy, D., Edwards, Z., Chittock, C., Grieve, A. and Young, S. J. (2004) ‘The dynamic risk assessment and management system: an assessment of immediate risk of violence for individuals with offending and challenging behaviour.’ Journal of Applied Research in Intellectual Disabilities 17(4), 267 274. Access Here.
Frize, C. M. (2015) The Assessment of Risk of General Recidivism in Offenders with an Intellectual Disability. Doctoral thesis. Sydney, Australia: University of Sydney. Access Here.
Lindsay, W. R., Steptoe, L. R., Hart, F., Miller, S., Macer, J. and McVicker, R. (2018) ‘The protective scale of the Armidilo S: The importance of forensic and clinical outcomes.’ Journal of Applied Research in Intellectual Disabilities, 1 8. Access Here
Craig, L. A., Browne, K. D. and Beech, A. R. (2008) Assessing Risk in Sex offenders: A Practitioner’s Guide. West Sussex, UK: John Wiley and Sons, Ltd. Access Here.
Lofthouse, R. E., Lindsay, W. R., Totsika, V., Hastings, R. P., Boer, D. P. and Haaven, J. L. (2013) ‘Prospective Dynamic Assessment of Risk of Sexual Reoffending in Individuals with an Intellectual Disability with a History of Sexual Offending Behaviour.’ Journal of Applied Research in Intellectual Disabilities 26(5), 394 403. Access Here
Blasingame, G. D., Boer, D. P., Guidry, L., Haaven, J., & Wilson, R. J. (2014). Assessment, treatment, and supervision of individuals with intellectual disabilities and problematic sexual behaviors. Beaverton, Oregon: Association for the Treatment of Sexual Abusers. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Pryboda, J., Tully, R. J. and Browne, K. D. (2015) ‘Is the Risk Matrix 2000 applicable to intellectually disabled sex offenders?’ Aggression and Violent Behavior 25(A), 184 190. Access Sindall,Here.
K. M. (2007) Assessing risk in a community intellectual disability sample: The Clinical Utility of the Dynamic Risk Assessment and Management System (DRAMS). Doctoral thesis. Leicester: University of Leicester. Access Here.
Cookman, C. L. (2010) ‘The utility of the assessment of risk management of ID individuals who offend sexually (ARMIDILO S) for a community based service.’ Dissertation Abstracts International, Massachusetts School of Professional Psychology. [Not accessible]
Steptoe, L. R., Lindsay, W. R., Murphy, L. and Young, S. J. (2008) ‘Construct validity, reliability and predictive validity of the dynamic risk assessment and management system (DRAMS) in offenders with intellectual disability.’ Legal and Criminological Psychology 13(2), 309 321. Access Here.
Levene, K. S., Augimeri, L. K., Pepler, D. J., Walsh, M. M., Webster, C. D. and Koegl, C. J. (2001) Early assessment risk list for girls, EARL 21G (Version 1). Toronto, ON: Earlscourt Child and Family Centre. [Not accessible]
Augimeri, L. K., Koegl, C. J., Levene, K. S. and Webster, C. D. (2005) ‘Early Assessment Risk Lists for Boys and Girls.’ In Grisso, T., Vincent, G. and Seagrave, D. (eds.). Mental health screening and assessment in juvenile justice. New York, NY: Guilford Press, 295 310. Access Here
Augimeri, L. K., Pepler, D., Walsh, M. M., Jiang, D. and Dassinger, C. (2010b). Aggressive and antisocial young children: Risk prediction, assessment and clinical risk management. Program Evaluation Report submitted to The Provincial Centre of Excellence for Child and Youth Mental Health at CHEO (Grant: # RG 976). Toronto, Ontario: Centre for Children Committing Offences and Program Development. Access Here
Koegl, C. J. (2011) High risk antisocial children. Predicting future criminal and health outcomes. Doctoral dissertation. Cambridge, UK: Institute of Criminology, University of Cambridge. [Not accessible]
Yuile, A. (2007) Developmental pathways of aggressive girls: A gender sensitive approach to risk assessment, intervention, and follow up. Doctoral dissertation. Toronto, Canada: York University. [Not accessible]
Augimeri, L., Walsh, M. and Donato, A. (2016) ‘SNAP® (Stop Now and Plan) and Future Criminal Outcomes: A Case for Intervention During the Middle Years.’ International Association of Forensic Mental Health Services, June 21st 23rd. New York: Fordham University. Access Here.
de Ruiter, C. and Augimeri, L. K. (2012) ‘Making delinquency prevention work with children and adolescents: From risk assessment to effective interventions.’ In C. Logan and L. Johnstone (Eds.), Managing clinical risk: A guide to effective practice. London, UK: Routledge, 199 223. Access Here
Campbell,FAM L. and Beech, A. (2018) ‘Do scores on the HCR 20 and FAM predict frequency of self harm in females within a secure psychiatric hospital?’ The Journal of Forensic Psychiatry & Psychology 29(6), 914 933. Access Here.
de Vogel, V. and de Vries Robbé, M. (2013) ‘Working with Women. Towards a more gender sensitive violence risk assessment.’ In L. Johnstone and C. Logan. (eds.) Managing Clinical Risk: a guide to effective practice. London: Routledge, 224 241. Access Here
Augimeri, L. K., Enebrink, P., Walsh, M. and Jiang, D. (2010a) ‘Gender Specific Childhood Risk Assessment Tools: Early Assessment Risk Lists for Boys (EARL 20B) and Girls (EARL 21G).’ In Otto, R. K. and Douglas, K. S. (eds.) Handbook of Violence Risk Assessment. London: Routledge, 43 62. Access Here.
EARL 21G
RATED page updated: August 2019 © Risk Management Authority 2019
RATED page updated: August 2019 © Risk Management Authority 2019
J., Blais, J. and Wilson, H. A. (2014) ‘A theoretically informed meta analysis of the risk for general and violent recidivism for mentally disordered offenders.’ Aggression and Violent Behavior 19(3), 278 287. Access Here.
Fazel, S., Wolf, A., Larsson, H., Lichtenstein, P., Mallett, S. and Fanshawe, T. R. (2017) ‘Identification of low risk of violent crime in severe mental illness with a clinical prediction tool (OxMIV): a derivation and validation study.’ Lancet Psychiatry 4(6), 461 468. Access Here
Warren, J., Chauhan, P. and Murrie, D. (1998) ‘Screening for Psychopathy Among Incarcerated Women: Psychometric Properties and Construct Validity of the Hare P SCAN.’ International Journal of Mental Health 4(2), 175 189. Access Here
Griswold, H., Green, D., Belfi, B., Grossi, L., Smith, J. and Otten, J. (2016) The Female Additional Manual (FAM): An investigation of its predictive validity among female defendants adjudicated not guilty by reason of insanity Paper Presented at the Annual Conference of the International Association for Forensic Mental Health Services (June), New York, NY.
Witt, K., van Dorn, R. and Fazel, S. (2013) ‘Risk factors for violence in psychosis: systematic review and meta regression analysis of 110 studies.’ PLoS One 8(2), e55942. Access Here.
de Vogel, V., Bruggeman, M. and Lancel, M. (2019) ‘Gender sensitive Violence Risk Assessment: Predictive Validity of Six Tools in Female Forensic Psychiatric Patients.’ Criminal Justice and Behavior [Online First]. Access Here
Brackett, R. E., Jackson, R. L. and Richards, H. J. (2008) ‘The Hare PSCAN and its relationship to psychopathy in a sample of civilly committed sexual offenders.’ The International Journal of Forensic Mental Health 7(1), 29 37. Access Here.
Elwood, C. E., Poythress, N. G. and Douglas, K. S. (2004) ‘Evaluation of the Hare P SCAN in a non clinical population.’ Personality and Individual Differences 36(4), 833 843. Access Here.
WRNA
de Vogel, V., de Vries Robbé, M., van Kalmthout, W. and Place, C. (2012) ‘Risk assessment of violent women: development of the “Female Additional Manual” (FAM).’ Tijdschrift voor Psychiatrie 54(4), 329 338. Access Here
P SCAN
Greig, D. G. (2014) Gender sensitive violence risk assessment; a preliminary investigation into the predictive accuracy and clinical utility of the FAM. Master of Arts dissertation. British Columbia, Canada: Simon Fraser University. Access Here.
de Vogel, V., de Vries Robbé, M., van Kalmthout, W. and Place, C. (2014) Female Additional Manual (FAM): Additional guidelines to the HCR 20V3 for assessing risk for violence in women. Utrecht, The Netherlands: Van der Hoeven Kliniek. [Not accessible]
Bonta,OxMIV
Negatsch, V., Voulgaris, A., Seidel, P., Roehle, R. and Opitz Welke, A. (2019) ‘Identifying Violent Behavior Using the Oxford Mental Illness and Violence Tool in a Psychiatric Ward of a German Prison Hospital.’ Frontiers in Psychiatry 10, 264. Access Here.
RATED page updated: August 2019 © Risk Management Authority 2019
Van Voorhis, P., Wright, E. M., Salisbury, E. and Bauman, A. (2010) ‘Women’s Risk Factors and Their Contribution to Existing Risk/Needs Assessment: The current status of a gender responsive supplement.’ Criminal Justice and Behaviour 37(3), 261 288. Access Here
Van Voorhis, P., Bauman, A. and Brushett, R. (2012) Revalidation of the women’s risk needs assessment institutional results: Pre release results. Washington, D. C.: National Institute of Corrections. Access Here.
Van Voorhis, P., Bauman, A. and Brushett, R. (2013) Revalidation of the women’s risk needs assessment institutional results: final report. Washington, D. C.: National Institute of Corrections. Access Here
Beppre, B. and Salisbury, E. (2016) ‘The Women’s Risk Needs Assessment: Putting Gender at the Forefront of Actuarial Risk Assessment.’ Penal Reform International. Access Here.
Wright, E. M., Salisbury, E. J. and Van Voorhis, P. (2007) ‘Predicting the Prison Misconducts of Women Offenders: The Importance of Gender Responsive Needs.’ Journal of Contemporary Criminal Justice 23(4), 310 340. Access Here.
Van Voorhis, P., Salisbury, E., Wright, E. and Bauman, A. (2008) Achieving Accurate Pictures of Risk and Identifying Gender Responsive Needs: Two New Assessments for Women Offenders. Washington, D.C.: United States Department of Justice: National Institute of Corrections. Access Here.
Brushett, R. A. (2013) Typologies of female offenders: a latent class analysis using the women’s risk needs assessment. Doctoral thesis. Cincinnati, OH: University of Cincinnati. Access Here.