Citation

Olver ME, Beggs Christofferson SM, Wong SC. Behav. Sci. Law 2015; 33(1): 92-110.

Affiliation

Department of Psychology, University of Canterbury, Christchurch, New Zealand.

Copyright

(Copyright © 2015, John Wiley and Sons)

DOI

10.1002/bsl.2159

PMID

25693953

Abstract

We examined the use of the clinically significant change (CSC) method with the Violence Risk Scale-Sexual Offender version (VRS-SO), and its implications for risk communication, in a combined sample of 945 treated sexual offenders from three international settings, followed up for a minimum 5 years post-release. The reliable change (RC) index was used to identify thresholds of clinically meaningful change and to create four CSC groups (already okay, recovered, improved, unchanged) based on VRS-SO dynamic scores and amount of change made. Outcome analyses demonstrated important CSC-group differences in 5-year rates of sexual and violent recidivism. However, when baseline risk was controlled via Cox regression survival analysis, the pattern and magnitude of CSC-group differences in sexual and violent recidivism changed to suggest that observed variation in recidivism base rates could be at least partly explained by pre-existing group differences in risk level. Implications for communication of risk-change information and applications to clinical practice are discussed. Copyright © 2015 John Wiley & Sons, Ltd.

Language: en