Structured professional judgement is the gold standard for forensic risk assessment proven by several studies. None of these studies however investigated the real strength of such a method, which lies in its function to guide clinical judgment by integrating the several evidence based risk factors in a professional process beyond arithmetical addition.

In our semi-prospective study (retrospective data-collection from 379 risk assessments from a forensic-psychiatric sample between 1989 and 2000 and a prospective follow-up with complete criminal records until 2007) we found the method of structured professional judgement by means of a catalogue of criteria being prospectively valid: With a median time at risk of 8.4 years the risk management (the process of risk assessment, court decision and sentencing respectively court ordered treatment) reduced both general as violent recidivism significantly.

37% of the subjects recidivated median 2.1 years after risk assessment. Recidivism rates (adjusted for time at risk ≥ 24 months) for homicide were 2.1%, 15.7% for all violent offenses, 7.1% for sexual offenses and 2.1% for child abuse.

None of the 56 subjects given a negative prediction (“no recidivism”) recidivated with serious offenses.

We will also present data about construct validity, CHAID-analysis for recidivism and comparison with PCL and HCR-20 scores.