Sorter M: Brief rating of aggression by children and adolescents (BRACHA): a reliability study. J Am Acadmodels having finite populations

This study evaluated the Brief Rating of Aggression by Children and Adolescents–Preliminary Version (BRACHA 0.8), an actuarial method of assessing the risk of aggressive behavior by hospitalized children and adolescents. Licensed psychiatric social workers used a 16-item questionnaire to assess all patients seen in the emergency department (ED) of a major urban children’s hospital. Over a six-month period, 418 patients (age range, 3.5–19.0 years) underwent psychiatric hospitalization after ED evaluation. The hospital nursing staff recorded the inpatients’ behavior, with the Overt Aggression Scale (OAS). Inpatients were deemed aggressive if, during the first six days of their hospital stay, they scored one or higher on any OAS subscale. We evaluated questionnaire properties, items, and demographic covariates (e.g., age, sex, and living situation) by using factor analyses, logistic regression models, and receiver operating characteristic (ROC) methods. A total of 292 aggressive acts were committed by 120 (29 % of 418) patients. Fourteen of the 16 items predicted (p .007) inpatient aggression and showed good internal consistency (Cronbach’s 0.837). Age was inversely related to probability of aggression and was incorporated into the final assessment instrument. Predictive power was comparable with other published risk assessment instruments (ROC areas of.75 for any aggression and.82 for aggression toward others). BRACHA 0.8 shows promise in rapidly assessing risk of inpatient aggression, but further studies are needed to establish the reliability and validity of the instrument