UNLABELLED: Evaluation of case-mix and resource consumption in pediatric intensive care units (ICU) is required. AIMS: This study describes the patterns of pediatric ICU resource consumption, determines the impact of primary clinical characteristics (particularly severity of illness) on resource utilization, and analyses medical efficiency with the frequency of inappropriate stays in French pediatric ICU. METHODS: Prospective study in nine French volunteer multidisciplinary pediatric ICUs from December 1993 to April 1994. Premature neonates were excluded. Resource consumption was expressed using the Omega system and length of stay, from which total Omega per admission, and average daily Omega (total Omega/length of stay) were obtained. RESULTS: Seven hundred and twelve patients were eligible. Twenty-five percent were full-term neonates, 30% infants (1 month to 1 year), and 45% children. Surgical patients constituted 22% of the population. A chronic disease was present in 45% of infants and children. Immunodeficiency was present in 10% of patients. The median length of stay was 4 days (range: 1-155). The mean Omega scores per admission were: total Omega = 92 +/- 124, Omega/day = 14 +/- 9. Sixty-four percent required mechanical ventilation and 37% during more than 2 days. Forty-two percent had a central venous access, and 23% an arterial line. The resource consumption was greater in non-survivors, surgical patients, neonates, and immunosuppressed patients. The mean PRISM score was 9 +/- 9. Mortality was 13%. The Omega/day and the PRISM score correlated. The frequency of inappropriate stays was 7.6% and accounted for 1.0% of the overall Omega activity. CONCLUSIONS: The results of this study can be used for interinstitutional comparison and a broader appraisal of pediatric intensive care. They illustrate the relationship between severity of illness and resource consumption. French pediatric ICU efficiency seems to be high, as compared to North American and Dutch results.