Selection of paediatric patients for intensive care.

1995 
Objectives. To determine characteristics of older infants and children admitted to the paediatric intensive care unit (PICU) at Johannesburg Hospital ; and to evaluate an admission score based on the PRISM score (an index of severity of illness) as a possible means for selection of patients for admission to the PICU. Design. Retrospective review of patient records and calculation of admission score from data. Setting. Paediatric intensive care unit at Johannesburg Hospital. Subjects. All patients older than 3 months of age admitted to the PICU from July 1993 to 31 March 1994. Results. There were 117 admissions during the study period with a mean age of 4,6 years. The mortality rate was 29,1%. The mean duration of ICU stay was 4,2 days. A review of requests for admission showed that over a 7-month period, 53 patients (> 3 months) could not be accommodated. In 71 patients with complete data, the admission score was significantly higher in non-survivors than survivors. The area under the ROC curve for predicting mortality from the admission score was 0,73 (SE 0,054). An admission score ≥ 16 predicted mortality with a sensitivity of 42% and a specificity of 98%. Conclusions. PICU facilities at Johannesburg Hospital are insufficient to meet the demand. An admission score based on the PRISM score could assist in the selection of patients for these limited PICU facilities.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    4
    Citations
    NaN
    KQI
    []