Differences between clinicians and researchers in assessing risk of violence in forensic psychiatric patients

Do clinicians and researchers differ in their violence risk assessment of the same patient? In this study, the Dutch version of the HCR-20 was coded by two independent researchers and two independent clinicians (treatment supervisor and group leader) for 60 patients admitted to a Dutch forensic psychiatric hospital. The aim of the study was threefold: (1) to establish the interrater reliability of the Dutch HCR-20; (2) to gain insight into differences between researchers and clinicians in coding the HCR-20; and (3) to examine the relationship between clinicians' feelings towards their patients and their risk judgment. Overall, the interrater reliability of the HCR-20 was good. The group leaders gave significantly lower HCR-20 scores than the researchers. There were no significant differences between the mean HCR-20 scores of treatment supervisors and researchers, but there was a significant difference in the interpretation of the scores: treatment supervisors had more 'low risk' judgments than researchers. Furthermore, it was found that feelings of clinicians towards their patients were associated with their risk judgment. Feelings of being controlled and manipulated by the patient were related to higher HCR-20 scores, whereas positive feelings (helpful, happy, relaxed) were related to lower risk judgments.