作者: Mary C. McLellan , Kimberlee Gauvreau , Jean A. Connor
DOI: 10.1111/CHD.12132
关键词:
摘要: Objective Most inpatient pediatric arrests are preventable by early recognition/treatment of deterioration. Children with cardiac disease have the highest arrest rates; however, warning scoring systems not been validated in this population. The objective study was to validate Cardiac Children's Hospital Early Warning Score (C-CHEWS) tool patients. associated escalation care algorithm directs: routine (score 0–2), increased assessment/intervention (3–4), or intensive unit (CICU) consult/transfer (≥5). Design Sensitivity and specificity were estimated based on retrospective review patients that experienced unplanned CICU transfer/arrest (n = 64) a comparison sample 248) admissions. previously Pediatric (PEWS) used for comparison. Patients' C-CHEWS scores compared calculated PEWS scores. Area under receiver operating characteristic (AUROC) curve measure discrimination. Results The AUROC 0.917 0.785 (P < .001). 0.902 vs. 0.782. sensitivity 96.9 ≥ 2), 79.7 (≥4), 67.2 (≥5) 81.1(≥2), 37.5 23.4 (≥5). 58.1 (≥2), 85.5 93.6 81.1 94.8 (≥4) 97.6 Lead time elevated (≥2) median 9.25 hours prior event PEWS, which 2.25 lead critical 2 0 .001). Conclusions C-CHEWS has excellent discrimination identify deterioration children performed significantly better than both as an ordinal variable when choosing cut points maximize AUROC. higher at all points.