Abstract P-481: DIFFICULT OF EVALUATING PEDIATRIC DELIRIUM BY CLINICAL NURSES

Abstract

Aims & Objectives: Pediatric delirium (PD) is associated with prolonged stay in pediatric intensive care unit (PICU) and higher mortality. As a consequence, to assess PD is important. We investigated whether PICU nurses could clinically identify the presence of PD in critically ill pediatric patients. Methods This is a prospective observing study in a single facility. The research team and clinical nurses simultaneously and independently assessed a convenience sample of PD for PICU patient. Central nerve system disease or patients receiving neuromuscular blockade were excluded. To assess PD, the research team used the Cornell Assessment of Pediatric Delirium (CAPD) score ≥ 9 and clinical nurses subjectively evaluated. We used the bootstrap method to measure the validity and reliability. Results A total of 100 paired PD assessments were completed for 32 patients with a median age of 23 months [interquartile range (IQR), 1–36]. Mechanical ventilation was necessary for 49% of the patients. Considering the CAPD as the reference standard, delirium occurred in 50% of the patient encounters. Delirium was poorly recognized by clinical nurses, sensitivity of 5.7% (95% CI, 1.5–9.8), specificity of 96% (95% CI, 84–97). Collectively, these data indicate that clinical nurses were difficult to subjectively evaluate PD. We recommend using validated delirium-screening instrument. Conclusions Collectively, these data indicate that clinical nurses were difficult to subjectively evaluate PD. We recommend using validated delirium-screening instrument.