Communicating the results of criterion referenced prediction measures: Risk categories for the Static-99R and Static-2002R sexual offender risk assessment tools.

Abstract

This article describes principles for developing risk category labels for criterion referenced prediction measures, and demonstrates their utility by creating new risk categories for the Static-99R and Static-2002R sexual offender risk assessment tools. Currently, risk assessments in corrections and forensic mental health are typically summarized in 1 of 3 words: low, moderate, or high. Although these risk labels have strong influence on decision makers, they are interpreted differently across settings, even among trained professionals. The current article provides a framework for standardizing risk communication by matching (a) the information contained in risk tools to (b) a broadly applicable classification of "riskiness" that is independent of any particular offender risk scale. We found that the new, common STATIC risk categories not only increase concordance of risk classification (from 51% to 72%)-they also allow evaluators to make the same inferences for offenders in the same category regardless of which instrument was used to assign category membership. More generally, we argue that the risk categories should be linked to the decisions at hand, and that risk communication can be improved by grounding these risk categories in evidence-based definitions. (PsycINFO Database Record

The Predictive Properties of Psychiatric Diagnoses, Dynamic Risk and Dynamic Risk Change Assessed by the VRS-SO in Forensically Admitted and Released Sexual Offenders

“…The Static-99 total score for the released sample was M = 4.66 (SD = 2.43) representing the risk category referenced as Level IVa or above average (6,13). The total score of the VRS-SO (pretreatment) was M = 46.93 (SD = 9.34), which was also Level IVa, above average (16).…”

“…Risk in sexual offenders can effectively be captured by several well validated risk assessment instruments (11); however, some of those instruments, such as the Static-99 and its revision, the Static-99R, are comprised only of static items (e.g., criminal history, offender and victim demographics; 12). Although such instruments typically reliably predict sexual recidivism and inform risk classification (13), their use with mentally ill offenders is limited given that their risk for sexual violence must be reduced to form a compelling argument for release.…”

The case against categorical risk estimates

“…Some risk categories are intended simply to be “evaluative labels” of the risk posed in a particular case (Hanson et al, , p. 585). Other risk categories include information about risk management.…”

“…The final risk category (IV‐b) was defined as the top 8% of the sample, and the fourth category (IV‐a) contained the individuals who were in between the final and middle categories. The team concluded “that the new STATIC categories have sufficiently improved conceptual coherence and have sufficient empirical support to replace the original categories in applied assessments” (Hanson et al, , p. 592).…”

Diagnosing Crime and Diagnosing Disease: Bias Reduction Strategies in the Forensic and Clinical Sciences

“…Communicating risk, especially about probability, is inherently difficult, and is made even more so when the risk being discussed carries such high individual and societal consequences . Some promising methods include mock juror studies and developing a “common language” of risk communication that minimizes misunderstandings and possible bias . Because this type of risk prediction is common across a broad range of forensic and clinical sciences, such research can assist experts from multiple disciplines in accurately and consistently communicating the uncertainty which is inherent in their opinions.…”