Untersuchung der prädikativen Validiät des Static-99 bei begutachteten Sexualstraftätern in der Schweiz

Abstract

Introduction: Various international validation studies have found a satisfactory to very good predictive validity for the Static-99, with AUC-values between .70 and .80. It is argued by the authors of this study that due to the static nature of the items assessed in the instrument it is primarily suited for first time psychiatric assessments and less for continuous assessment of offenders' risk dispositions. So far no validation studies of the Static-99 were made on samples of first time psychiatrically assessed offenders in Switzerland. Method: The Static-99 scores of sixty-three sex offenders in Switzerland were assessed by using data from their psychiatric assessments. At no point of the study was there personal contact with the offenders. Recidivism was operationalized as reconviction assessed from penal records. Any reconvictions as well as reconvictions due to sexual and violent offenses were coded and analyzed separately. The follow-up-period was between 7-32 years. Results: The Static-99 risk categories showed satisfactory predictive validity for reconvictions due to any offense (AUC = .64) and good predictive validity for reconvictions due to violent offenses (AUC = .81). They were however, unable to predict sexual recidivism in our population. Discussion: The Static-99 is a valid instrument to predict recidivism also in Swiss offender populations. In this sample of psychiatrically assessed sex offenders the Static-99 was however not able to predict sexual recidivism contrary not only to other international studies but also to a study on Swiss sex and violent offenders administered by the Zurich office of penal corrections. This finding suggests that risk assessment instruments are only valid for specific offender populations and that they can not be applied to other offender groups without further evaluation.