Abstract
Performance of a 10 bed paediatric intensive care unit (PICU) in a university children's hospital was assessed in a prospective cohort study of unselected, consequtive admissions. Collected data included primary clinical specialty, PRISM score derived mortality risk, administration of PICU-dependent therapy and vital status at discharge. Effectiveness of care was determined by comparing severity of illness based predicted mortality with vital status at discharge. Efficiency was determined by the administration of at least one PICU-dependent therapy or mortality risk exceeding 1%. 593 patients were included, PICU mortality was 8.4%. The overall performance of the PRISM score based predictive model was found to be well (goodness-of-fit test X2(5) = 5.49, p=0.33; area under ROC-curve 0.92), indicating good effectiveness of care. Overall 489 of 593 (82.5%) admissions were efficient, as were 2393 of 3130 (76.5%) PICU days. In cardiovascular surgery patients 800 of 927 (86.3%) PICU days were efficient, in non-surgical patients 1421/1855 (76.6%), and in other surgery patients 172/347 (49.6%).
In conclusion, performance of paediatric intensive care could be determined using objective criteria (mortality risk and administration of PICU-dependent therapy), allowing inter-institutional comparative assessment of quality of care in the future.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Gemke, R., Bonsel, G. & Van Vught, A. 362 Assessing performance of paediatric intensive care. Pediatr Res 36, 63 (1994). https://doi.org/10.1203/00006450-199407000-00362
Issue Date:
DOI: https://doi.org/10.1203/00006450-199407000-00362