Too risky to use, or too risky not to? Lessons learned from over 30 years of research on forensic risk assessment with Indigenous persons

Indigenous peoples are overrepresented in correctional systems internationally, reflecting a history of systemic racism and colonial oppression, and the practice of risk assessment with this population has been a focus of legal and sociopolitical controversy. We conducted a systematic review and meta-analysis of the risk assessment literature comparing Indigenous and non-Indigenous (White majority) groups. We retrieved 91 studies featuring 22 risk tools and 15 risk/need/cultural domains (N = 59,693, Indigenous; N = 237,729, non-Indigenous/White) and four documents identifying culturally relevant factors. Most measures demonstrated moderate predictive validity but often had significant ethnoracial differences, particularly for static measures. The Service Planning Instrument/Youth Assessment Screening Inventory, Level of Service Inventory youth variants, Psychopathy Checklist-Revised and Youth Version, and the Violence Risk Scale and its Sexual Offense version had the strongest predictive validity and least ethnoracial discrepancy. The Static Factors Assessment and Dynamic Factors Identification and Analysis-Revised had the weakest predictive validity. For Indigenous persons, the strongest individual predictors of recidivism were low education/employment, substance abuse, antisocial pattern, and poor community functioning, while mitigating factors that predicted decreased recidivism were measures of risk change (i.e., from culturally integrated programs combining mainstream and traditional healing approaches), cultural engagement/connectedness, and protective factors. In practice, static measures need to be supplemented with dynamic ones, and assessors should select measures with at least moderate predictive validity and ideally the least ethnoracial bias. These conclusions are tempered by the quantity and quality of the literature coupled with the circumstance that some study authors have coauthored tools in this review. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

The Effect of Missing Item Data on the Relative Predictive Accuracy of Correctional Risk Assessment Tools

Missing data are pervasive in risk assessment but their impact on predictive accuracy has largely been unexplored. Common techniques for handling missing risk data include summing available items or proration; however, multiple imputation is a more defensible approach that has not been methodically tested against these simpler techniques. We compared the validity of these three missing data techniques across six conditions using STABLE-2007 (N = 4,286) and SARA-V2 (N = 455) assessments from men on community supervision in Canada. Condition 1 was the observed data (low missingness), and Conditions 2 to 6 were generated missing data conditions, whereby 1% to 50% of items per case were randomly deleted in 10% increments. Relative predictive accuracy was unaffected by missing data, and simpler techniques performed just as well as multiple imputation, but summed totals underestimated absolute risk. The current study therefore provides empirical justification for using proration when data are missing within a sample.

Cross-Cultural Validity of Sexual Recidivism Risk Assessments Using Static-99R, STABLE-2007, and the VRS-SO

PURPOSE OF REVIEW

The overrepresentation of certain racial/ethnic groups in criminal legal systems raises concerns about the cross-cultural application of risk assessment tools. We provide a framework for conceptualizing and measuring racial bias/fairness and review research for three tools assessing risk of sexual recidivism: Static-99R, STABLE-2007, and VRS-SO.

RECENT FINDINGS

Most cross-cultural research examines Static-99R and generally supports its use with Black, White, Hispanic, and Asian men. Preliminary research also supports STABLE-2007 with Asian men. Findings are most concerning for Indigenous men, where Static-99R and STABLE-2007 significantly predict sexual recidivism, but with significantly and meaningfully lower accuracy compared to White men. For the VRS-SO and the combined Static-99R/STABLE-2007 risk levels, predictive accuracy was not significantly lower for Indigenous men, for which we discuss several possible explanations. We offer considerations for risk scale selection with Indigenous men and highlight recent guidance produced for cross-cultural risk assessment.

ACUTE-2007 and STABLE-2007 predict recidivism for men adjudicated for child sexual exploitation material offending

OBJECTIVE

Risk assessment is essential to effective correctional practice. For individuals with contact sexual offenses, many risk tools are available. There are fewer options, however, for individuals whose sexual offending exclusively involves child sexual exploitation materials (CSEM; legally referred to in Canada and the United States as child pornography).

HYPOTHESES

The present study examined the predictive validity of the ACUTE-2007 and STABLE-2007 sexual recidivism risk tools among men with CSEM offenses. We expected these tools to show moderate predictive validity across study groups.

METHOD

We compared the scales' discrimination and calibration across three groups: (a) 1,042 men with contact sexual offenses against children (baseline comparison), (b) 228 men with exclusive CSEM offending (no contact sexual offenses), and (c) 80 men with both contact sexual offenses and CSEM offenses.

RESULTS

We found that the ACUTE-2007 and STABLE-2007 total scores and items had comparable (and often better) discrimination for men with CSEM offending compared with contact sexual offending against children in the prediction of any sexual recidivism, violent recidivism, and any recidivism. Calibration analyses indicated that the overall sexual recidivism rates for the median ACUTE-2007 and STABLE-2007 scores were similar for men with exclusive CSEM offenses compared with men with any contact offending against children. Almost all of the sexual recidivism for the CSEM-exclusive group involved further CSEM offenses.

CONCLUSIONS

This study supports the use of these tools to rank-order men with CSEM offending in terms of their risk of reoffending and to help direct treatment and management efforts. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Same Score, Different Audience, Different Message: Perceptions of Sex Offense Risk Depend on Static-99R Risk Level and Personality Factors of the Recipient

There are multiple ways to report risk scale results. Varela et al. (2014) found that Static-99R results were interpreted differently by prospective jurors based on risk level (high vs low) and an interaction between risk level and risk communication format (categorical, absolute estimate, and risk ratio). We adapted and extended Varela et al.'s (2014) study using updated Static-99R norms, recruiting a population-wide sample (n = 166), and adding variables assessing the personality factors 'cognitive motivation' (i.e., need for cognition) and 'attitudinal affect' (i.e., attitudes toward sex offenders, authoritarianism). We found a main effect of risk level and no effect of either communication format or the interaction between the two. Adding the personality variables increased explained variance from 9% to 34%, suggesting risk perception may be more about the personality of the person receiving the information than the information itself. We also found an interaction between attitudes toward sex offenders and risk level. Our results suggest risk perception might be better understood if personality factors are considered, particularly attitudes toward sex offenders. Because biases/personality of the person receiving the information are unknown in real world settings we argue that sharing multiple methods for communicating risk might be best and more inclusive.

Emotional Congruence with Children: An Empirical Examination of Different Models in Men with a History of Sexually Offending Against Children

Emotional congruence with children (ECWC) is a psychologically meaningful risk factor for sexual offending against children (SOC). Based on previous research and theory, three models have been proposed to explain ECWC: Blockage, Sexual Domain, and Psychological Immaturity. Using structural equation modelling in a routine correctional sample of men adjudicated for sexual offences (n = 983), we found little support for all three of these models. Instead, we found that atypical sexual interests, alone, best explained ECWC, with a moderate relationship to ECWC. Using the predictors associated with each of the three models of ECWC, we identified three classes of men with a history of SOC who are high in ECWC using latent class analyses (n = 377). These three classes generally did not replicate the three models of ECWC. We instead propose three subgroups of men with histories of SOC who are high in ECWC, characterized respectively by: relationship deficits; youth and loneliness; and high sexual and general criminality. High levels of ECWC are predictive of a higher risk of sexual recidivism, regardless of class association; however, these subgroups are differentially at risk for some types of recidivism. Our findings suggest that ECWC is a multi-faceted construct, which is still not well understood.

Prevalence, Correlates, and Sequelae of Child Sexual Abuse (CSA) among Indigenous Canadians: Intersections of Ethnicity, Gender, and Socioeconomic Status

Child sexual abuse (CSA) is a severe and concerning public-health problem globally, but some children are at higher risk of experiencing it. The harms caused by colonization and particularly the inter-generational legacy of residential schools would presumably increase the vulnerability of Indigenous children in former British colonies. Among 282 Indigenous participants in Canada recruited from Prime Panels, CSA was reported by 35% of boys, 50% of girls, and 57% of trans and gender non-conforming participants. These rates are substantially higher than global meta-analytic estimates (7.6% of boys and 18.0% of girls). There was evidence of intersectionality based on socioeconomic status. CSA was associated with a variety of other indicators of negative childhood experiences and significantly predicted numerous negative outcomes in adulthood, including mental-health issues (e.g., PTSD), unemployment, and criminal legal-system involvement. Sexual abuse of Indigenous Canadian children is a public-health crisis, and layers of marginalization (e.g., gender, social class) exacerbate this risk. Trauma-informed services to address the harms of colonization are severely needed, in line with recommendations from Canada's Truth and Reconciliation Commission.

Predictive accuracy of Static-99R across different racial/ethnic groups: A meta-analysis

OBJECTIVE

The overrepresentation of numerous racial/ethnic groups in the criminal legal system warrants examination of the cross-cultural applicability of risk assessment tools. Static-99R is a tool used in diverse countries to assess sexual recidivism risk. We conducted a meta-analysis on the predictive accuracy of Static-99R across different racial/ethnic groups.

HYPOTHESES

No hypotheses were made regarding discrimination, given that past research could support hypotheses of differential or equivalent accuracy. We hypothesized that Indigenous individuals would score higher on Static-99R than non-Indigenous or White individuals.

METHOD

Our search identified 18 eligible documents (from 17 distinct studies) with 41 nonoverlapping effect sizes. These 17 studies examined the predictive accuracy of Static-99R with racially/ethnically diverse men charged with or convicted of sexually motivated offenses. We report analyses using both fixed-effect and random-effects meta-analysis.

RESULTS

Indigenous and Black individuals scored significantly higher on Static-99R than their non-Indigenous or White counterparts, with small effect sizes. For discrimination, area under the curve (AUC) values were generally moderate-to-large and statistically significant for all groups in both fixed-effect and random-effects analyses. Within-study subgroup analyses indicated significantly lower accuracy for Indigenous and Hispanic individuals compared with White/non-Indigenous samples (though for Hispanic individuals, this finding was significant only in the fixed-effect analyses). No statistically significant differences in accuracy were found between White and Black individuals. Static-99R significantly predicted recidivism with large effect sizes across two samples of Asian individuals. Two studies supported calibration across Black, White, and Hispanic individuals. Two studies examining calibration of Static-99R for Indigenous individuals had mixed findings.

CONCLUSIONS

Given a small number of studies and limitations with both the fixed- and random-effects analyses, readers should interpret findings regarding Hispanic individuals with caution. The analyses clearly found significant but lower accuracy for Static-99R with Indigenous individuals. Potential reasons for this differential accuracy are discussed, along with limitations of the meta-analysis and suggestions for research and practice. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Improving Risk Communication: Developing Risk Ratios for the VRAG-R

We developed a set of risk ratios for the Violence Risk Appraisal Guide-Revised (VRAG-R) to broaden the range of risk communication options available when using this tool and to provide information needed for future efforts to apply The Council of State Governments Justice Center's standardized five-level risk framework to the scale. A slightly reduced version of the VRAG-R normative data set was used for the analyses (N = 1,238). Contrary to previous research developing risk ratios, logistic regression provided a more accurate estimate of observed violent recidivism rates than Cox regression for both total VRAG-R scores and VRAG-R decile bins. Further analyses indicated the relationship between the VRAG-R and violent recidivism was consistent over a 15-year follow-up period. Due to the difficulties with interpreting odds ratios, the final risk ratios were computed using rate ratios derived from a logistic regression model using a 5-year fixed follow-up period. These risk ratios, and templates for how the ratios might be used in an assessment report, are presented in the appendices.

Using Graphs in Sexual Violence Risk Communication: Benefits May Depend on the Risk Metric

Actuarial scales provide a relatively objective and reliable assessment of individuals' risk of recidivism. Recent research has explored how graphs can improve quantitative risk communication. We tested whether graphs can improve understanding and perception of sexual violence risk when matched with risk metric. Participants (N = 676) were recruited from Amazon's MTurk platform and read a brief description of a man convicted of a sexual offense, including results of a fictional sexual recidivism risk scale. In Study 1, absolute risk of recidivism enabled participants to distinguish between individuals with relatively high and low risk of sexual recidivism. In Study 2, this distinction was enhanced by adding a graph, especially when percentiles were communicated. Risk ratios increased perceived risk. Objective numeracy increased understanding and reduced perceived risk. We recommend that risk communication assumes limited statistical numeracy, and further research with practitioners to test the effect of graphs and risk metrics on forensic/judicial decisions.

The Static-99R Is Not Valid For Women: Predictive Validity in 739 Females Who Have Sexually Offended

Although many instruments have been validated to assess risk of sexual recidivism among men, no similar tool exists for women who have sexually offended. As a result, some jurisdictions use male-based instruments to assess women despite the lack of validation research examining the predictive utility for this subgroup. This study examined the utility of the Static-99R in predicting sexual recidivism among women. Based on a sample of 739 women convicted of sexual offenses in Texas, findings show that the total score was not significantly associated with sexual recidivism. When looking at individual items, other than a history of prior sexual offenses and noncontact sexual offenses, no item of the Static-99R was significantly associated with sexual recidivism among women. Furthermore, only three items (female victims/solo offender, 4+ sentencing dates, and having not lived with a romantic other for 2+ years) were significantly associated with nonsexual recidivism. These findings indicate that the Static-99R is not suitable to assess risk of recidivism among women convicted of sexual offenses.

Estimating the probability of sexual recidivism among men charged or convicted of sexual offences: Evidence-based guidance for applied evaluators

Risk assessment is routinely applied in forensic decision-making. Although relative risk information from risk scales is robust across diverse samples and settings, estimates of the absolute probability of sexual recidivism are not. Nonetheless, absolute recidivism estimates are still necessary in some evaluations. This paper summarizes research and offers guidance on evidence-based practices for assessing the probability of recidivism, organized largely around questions commonly asked in court. Overall, estimating the probability of sexual recidivism is difficult and should be undertaken with humility and circumspection. That being said, research favours empirical-actuarial risk tools for this task, more structured scales, and the use of multiple scales. Professional overrides of risk scale results should not be used under any circumstances. Paradoxically, however, professional judgement is still required in some circumstances. Risk scales do not consider all relevant risk factors, but the added value of external risk factors reaches a point of diminishing returns and may or may not be incremental (or worse, can degrade accuracy). There are reasons actuarial risk scales may both underestimate recidivism (e.g., undetected offending, short follow-ups) and overestimate recidivism (e.g., inclusion of sex offences not of interest in some referral questions, data on declining crime and recidivism rates, newer studies demonstrating overestimation of recidivism). Given all these considerations and the need for humility, in the absence of exceptional circumstances, I would not deviate too far from empirical estimates.

Field validity of Static-99R and STABLE-2007 with 4,433 men serving sentences for sexual offences in British Columbia: New findings and meta-analysis

Many forensic assessment measures are developed and validated under research conditions but applied in the field, where professionals or paraprofessionals have varied training, unknown fidelity to administration procedures, and contextual pressures related to their institutions or legal system. Yet few studies examine the generalizability of psychometric properties of these scales as actually applied in field settings. This study examined 4,433 individuals assessed by probation officers on the Static-99R or STABLE-2007 sexual recidivism risk scales in British Columbia, Canada. Sexual, violent, and any recidivism were examined. Static-99R and STABLE-2007 had moderate accuracy in discriminating recidivists from non-recidivists, and both scales added incrementally in predicting all three outcomes (with Static-99R demonstrating higher accuracy). Organizing the items into constructs, sexual criminality, general criminality, and youthful stranger aggression incrementally predicted all three outcomes. For violent and any recidivism, the incremental effect of sexual criminality was in the negative direction (i.e., high sexual criminality was associated with relatively lower rates of violent and any recidivism). Calibration analyses indicated that recidivism rates were lower than what would be predicted by the norms for the scales. The current study also presented a meta-analysis of 15 field validity studies of Static-99R and 4 field validity studies of STABLE-2007. Results of the current study and meta-analysis support the field application of Static-99R and STABLE-2007, while emphasizing the importance of training and proper implementation. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

A Validation Study of the Child Pornography Offender Risk Tool (CPORT)

The Child Pornography Offender Risk Tool (CPORT) is a seven-item structured tool to assess the likelihood of future sexual offending over a 5-year fixed follow-up. The current study examined 5-year fixed follow-up data (15% any new sexual offense, 9% any new child pornography offense) for a validation sample of 80 men convicted of child pornography offense(s). Although statistical power was low, results were comparable with the development sample: The CPORT had slightly lower predictive accuracy for sexual recidivism for the overall group (area under the curve [AUC] = .70 vs. .74), but these values were not significantly different. Combining the development and validation samples, the CPORT predicted any sexual recidivism (AUC = .72) and child pornography recidivism specifically (AUC = .74), with similar accuracies. CPORT was also significantly predictive of these outcomes for the child pornography offenders with no known contact offenses. Strengths and weaknesses of incorporating CPORT into applied risk assessments are discussed.

Sex Offender Risk Assessment: Where Are We and Where Are We Going?

PURPOSE OF REVIEW

Risk assessment is one of the most ubiquitous tasks in the criminal justice system, informing virtually every decision made about offenders. This review, intended for researchers and practitioners, outlines some of the most important recent advances, emerging issues, and recommendations in sex offender risk assessment.

RECENT FINDINGS

The underlying nature and purpose of risk scales is reviewed, with implications for how we should evaluate them. Limits of recidivism probability estimates are discussed, and efforts to advance a common language for describing risk levels are highlighted. Advances in risk communication and field validity are summarized. The utility of protective risk factors in risk assessments is debated. Emerging areas in assessing offender change and assessments with child pornography offenders are discussed. Despite critical advances in the last few years, there are still important gaps in knowledge, particularly for risk communication, field implementation, offender change, and child pornography offenders.

Emotional Congruence With Children Is Associated With Sexual Deviancy in Sexual Offenders Against Children

Emotional congruence with children is a psychologically meaningful risk factor for sexual offending against children. The present study examines the correlates of emotional congruence with children in a sample of 424 adult male sexual offenders who started a period of community supervision in Canada, Alaska, and Iowa between 2001 and 2005. Consistent with previous work, we found sexual offenders against children high in emotional congruence with children were more likely to be sexually deviant, have poor sexual self-regulation, experience social loneliness, and have more distorted cognitions about sex with children. Overall, our findings are most consistent with a sexual deviancy model, with some support for a blockage model.

Field validity of Static-99/R scores in a statewide sample of 34,687 convicted sexual offenders

The Static-99 (and revision, the Static-99R) reflect the most researched and widely used approach to sex offender risk assessment. Because the measure is so widely applied in jurisdictions beyond those on which it was developed, it becomes crucial to examine its field validity and the degree to which published norms and recidivism rates apply to other jurisdictions. We present a new and greatly expanded field study of the predictive validity (M = 5.23 years follow-up) of the Static-99 as applied system-wide in Texas (N = 34,687). Results revealed stronger predictive validity than a prior Texas field study, especially among offenders scored after the release of an updated scoring manual in 2003 (AUC = .66 to .67, d = .65 to .69), when field reliability was also stronger. But calibration analyses revealed that the Static-99R routine sample norms led to a significant overestimation of risk in Texas, especially for offenders with scores ranging from 1 to 5. We used logistic regression to develop local Texas recidivism norms (with confidence intervals) for Static-99R scores. Overall, findings highlight the importance of revisiting and updating field study findings, and the potential benefits of using statewide data to develop local norms. (PsycINFO Database Record