Convergent and predictive validity of the Sexual Violence Risk-20 and risk for sexual violence protocol with older sexual offenders

Structured professional judgement (SPJ) guidelines for assessing risk of sexual offending have existed for more than two decades, yet the literature investigating the validity of these tools remains relatively limited. Recent research has shown that actuarial sexual offending risk assessment tools should be adjusted to reflect reduced risk as offenders age (Helmus et al., 2012), leading to questions about the validity of extant SPJ guidelines for older people who sexually offend. The current study investigated the reliability and predictive validity of the Sexual Violence Risk-20 (SVR-20) and the Risk for Sexual Violence Protocol (RSVP) with a sample of 95 Australian individuals aged 50 years or over who had sexually offended. Sixteen (16.84%) participants sexually recidivated over a mean follow-up period of 9.07 years. Contrary to results with younger offenders, structured judgements using the SVR-20 or RSVP did not significantly differentiate recidivists from non-recidivists in this sample (AUC = .57). However, the total scores for both instruments were predictive at particular time periods. Certain risk factors were potentially important and accounted for unique variance that was not accounted for by actuarial instruments. The implications and limitations of this study are discussed, as well as recommendations for future research and practice.