Development and Validation of Early Warning Criteria to Identify Escalated Care Events in Neonatal Intensive Care Unit Patients

Abstract

Objective This study aimed to identify and validate the diagnostic utility of a set of clinical and laboratory criteria (early warning criteria [EWC]) that portend a clinical deterioration event (escalated care event [ECE]) in neonatal intensive care unit (NICU) patients. Study Design Using the RAND appropriateness method, we first established a consensus on seven ECE, that is, events that require additional monitoring, treatment, or stay in the NICU or that were associated with morbidity. We then established consensus on EWC that could portend an ECE from an initial set of 32 potential EWC items to a final set of 10 items. The occurrence and nonoccurrence of EWC and ECE were prospectively identified and tracked over 9 weeks. Results Among 170 NICU patients studied (2,502 patient-days), the frequency of an EWC was 53 per 1,000 patient-days. Of these patients, 41% had an EWC and 16% had an ECE. An EWC was followed by an ECE within 72 hours, 37% of the time, and within a median time interval of 113 minutes. The sensitivity, specificity, positive predictive values, and negative predictive values of EWC in identifying an ECE were 0.96, 0.69, 0.37, and 0.99, respectively. Conclusion A simple bedside NICU-specific EWC identifies neonates likely to develop ECEs in the NICU.