A Prospective Multisite Examination of Dynamic Sexual Violence Risk: Extension and Update to Olver, Nicholaichuk, Kingston, and Wong (2014)

Objective: The present study is an extension and update of Olver, Nicholaichuk, Kingston, and Wong's (2014) prospective multisite examination of sexual violence risk and treatment change on a large federal Canadian sample of 570 treated sexual offenders rated pretreatment and posttreatment on the Violence Risk Scale-Sexual Offense version (VRS-SO). Method: The present study featured the clinical application of a risk assessment and treatment planning tool, the VRS-SO, with recidivism outcome data updated by 4 years to a total of 10.2 years. VRS-SO change scores, representing reductions in sexual violence risk from treatment or other change agents (e.g., aging), were reanalyzed in terms of their associations with community sexual, violent, and general recidivism postrelease. Results: Recidivism base rates increased with the concordant increase in follow-up time. VRS-SO risk scores significantly predicted all recidivism outcomes, whereas change scores were consistently associated with decreases in sexual and violent recidivism after controlling for baseline risk through bivariate (d = −.24 to −.61) and Cox regression survival analyses (eB = .878 to .938). Conclusions: The results affirm the dynamic nature of sexual violence risk and demonstrate that structured assessments of change, linked to participation in sexual offense specific treatment, to be associated with decreases in future sexual offending as well as other recidivism outcomes. What is the public health significance of this article? The research demonstrates that structured assessments of change in sexual violence risk, completed by sexual offender treatment service providers in the field using a validated sexual offense risk assessment tool, are associated with reductions in sexual offense recidivism as well as other forms of recidivism. The results further underscore the importance of long-term follow-up and comprehensive recidivism data to maximize the accuracy and forensic clinical utility of prediction findings.