Evaluation of the pediatric risk of admission score in a pediatric emergency department

Study objectives: The pediatric risk of admission (PRISA) score was developed to predict the risk for hospitalization for pediatric emergency department patients. We sought to evaluate prospectively the predictive value of the PRISA score with respect to the risk for hospitalization in a pediatric ED. Methods: A prospective cohort study was conducted in a pediatric tertiary center ED. From November 1, 2000, to October 31, 2001, 3 periods of 8 hours each were randomly chosen monthly. During these periods, all patients triaged to the ED were evaluated. Data collection was performed by an investigator uninvolved in the patients' treatment. Data were recorded before the decision regarding hospitalization was made. Odds ratios for the risk of hospitalization related to individual criteria and PRISA scores were calculated. Discrimination and calibration of the score were assessed. Results: During the study periods, 1,930 patients were evaluated. Among these, 203 hospitalizations were observed, and the PRISA score predicted 235. The goodness-of-fit test demonstrated that the score had good predictive ability (chi(2) =28.15; P=38). Receiver operating characteristic curve analysis confirmed the latter findings (area under the curve 0.79 [95% confidence interval 0.72 to 0.86]). Some individual criteria of the score did not significantly predict risk for admission. Conclusion: The PRISA score is a good predictor of the risk for hospitalization in a pediatric ED. It seems more accurate for the sicker patients. Some variables of the score could be deleted or modified to optimize its accuracy.