Association between short-term structured risk assessment outcomes and seclusion.

Research findings indicate that the symptoms and behaviour of acute psychiatric patients can fluctuate drastically within hours, and that structured daily risk assessments can reduce the risk of aggressive incidents and the duration of seclusion. The aim of this study was to investigate the validity of two structured observation tools, the Brøset Violence Checklist (BVC) and the Kennedy Axis V), as an aid in seclusion-related clinical decision-making. In this study, 7403 day-to-day risk assessments were collected over 10 725 admission days (72% of the maximum number of structured assessments). A total of 7055 daily assessment scores from 301 acute psychiatric patients were used for the multilevel analysis. The sample demonstrated that dynamic and static factors were related to seclusion. Dynamic factors included dysfunctional scores on the item 'confusion' of the Brøset Violence Checklist, and psychological impairment and impairment of social skills on the Kennedy Axis V. Static factors included non-Western descent, male sex, age less than 35 years, unmarried, and to some extent, a personality disorder. McFadden's pseudo R(2) value showed that most of the final model was related to the dynamic factors. We concluded that the incorporation of the BVC and the Kennedy Axis V into standard practice was helpful in identifying patients at high risk of seclusion.