Validity of the updated pediatric risk of mortality score (PRISM III) in predicting the probability of mortality in a pediatric intensive care unit

This study assessed the validity of the PRISM III scoring system in accurately predicting the probability of mortality in a pediatric intensive care unit (PICU) in Taiwan. We collected data from consecutive patients admitted to our PICU during a one-year period from November 1999 through October 2000. Our PICU had 412 admissions including 385 patients (202 boys and 183 girls) with an average age of 53.9±58.2 months (range: 1 month to 18 years). The most common reason for admission was respiratory failure (26.2%). Almost 60% of the patients had at least one underlying chronic disease. The average duration of stay was 4.52±8.43 days (range 1 day to 81 days). The average PRISM III score was 5.06±6.95 (range 0-44). The overall mortality rate in the PICU was 8.17%. The mortality rate was not significantly different from the predicted rate (7.56%) (p=0.65). The efficiency of care was 32.5% and the standardized PICU length of stay ratio (SLOSR) was 1.33. The SLOSR ratio was significantly higher than the predicted rate (p value < 0.001). The PRISM III score was valid for assessing mortality risk of PICU patients in this hospital in Taiwan. However, the SLOSR ratio obtained in this study was higher than predicted. Differences in the patterns of practicing medicine may play an important role in observed PICU length of stay. Special caution is needed in adopting a severity of illness scoring system to assess performance of care, particularly in contexts different from the ones in which the instrument was originally developed. Further study including more pediatric intensive care units and other regions would enable greater generalization of the validity of this instrument.