To study the impact of trauma patients on Pediatric Intensive Care Units (PICUs), 164 trauma patients' data from 1,075 consecutive admissions to five PICUs were reviewed. Resource use (Therapeutic Intervention Scoring System [TISS] points) and mortality risks (Physiologic Stability Index [PSI] and Pediatric Risk of Mortality [PRISM] scores) were obtained daily for all patients. Trauma patients constituted 15.2% of all PICU patients, and used 14.9% of patient care days and 14.5% of TISS points. Efficiency of trauma patient care was 75% overall compared to 79% overall for nontrauma patients (p less than 0.001). Trauma patient mortality was 9.8%. Tests for goodness of fit showed the PSI and PRISM scores to be accurate outcome predictors for trauma patients (PSI: chi 2 (4) = 2.852, p greater than 0.50; PRISM: chi 2 (4) = 1.216, p greater than 0.50). Trauma patients are a minority of PICU patients and deaths. Their resource use is proportional to their numbers, although less efficient than for nontrauma patients. PSI and PRISM are accurate mortality risk predictors for trauma patients.