A Pilot Study on the Predictive Validity of Prognostic Tools in a Risk Group of Assessed Young Offenders Abstract: Background: This pilot study examined associations between established risk assessment instruments (PCL-YV/R, SAVRY, HCR-20 V3) and later criminal outcomes in a sample of 20 justice-referred adolescents, comparing those who reoffended in adulthood with those who did not. In addition to instrument scores, we assessed psychiatric diagnoses and social/criminological factors. Methods: We conducted a matched case-control and longitudinal analysis using data from forensic psychiatric/psychological evaluations in adolescence and, for the recidivist group, follow-up assessments in adulthood. We tested group differences at both the scale and domain levels. Results: Total instrument scores showed no significant group differences, though higher SAVRY total scores were linked with a greater frequency of subsequent violent convictions. In addition, a higher risk management factor score on the HCR-20 V3 was associated with later delinquency, indicating an increased need for support and intervention. This finding can be interpreted both as reflecting individual risk constellations and as potentially indicating insufficient structural conditions. Psychiatric diagnoses as well as social and criminological risk factors were highly prevalent in both groups. Conduct disorders occurred more frequently in the recidivism group, which also committed more violent offenses - a finding that underscores the importance of targeted tertiary preventive measures. Over time, risk assessments remained stable or increased slightly. Larger studies are needed to further examine the associations between risk assessment scores and later delinquency in high-risk groups of juvenile offenders.