1. Beyond Reporting Early Warning Score Sensitivity: The Temporal Relationship and Clinical Relevance of 'True Positive' Alerts that Precede Critical Deterioration
- Author
-
Christopher P. Bonafide, Meredith C. Winter, and Sherri Kubis
- Subjects
Male ,medicine.medical_specialty ,Adolescent ,Leadership and Management ,Cross-sectional study ,MEDLINE ,Vital signs ,Assessment and Diagnosis ,Intensive Care Units, Pediatric ,law.invention ,03 medical and health sciences ,0302 clinical medicine ,law ,030225 pediatrics ,Medicine ,Humans ,Clinical significance ,030212 general & internal medicine ,Child ,Care Planning ,Retrospective Studies ,Clinical Deterioration ,business.industry ,Vital Signs ,Health Policy ,Infant, Newborn ,Infant ,Retrospective cohort study ,General Medicine ,Early warning score ,Hospitals, Pediatric ,Intensive care unit ,Cross-Sectional Studies ,Early Warning Score ,Child, Preschool ,Emergency medicine ,Etiology ,Fundamentals and skills ,Female ,business ,Child, Hospitalized - Abstract
Background Clinical deterioration is difficult to detect in hospitalized children. The pediatric Rothman Index (pRI) is an early warning score that incorporates vital signs, laboratory studies, and nursing assessments to generate deterioration alerts. Objectives (1) Evaluate the timing of pRI alerts and clinicians recognizing deterioration or escalating care prior to critical deterioration events (CDEs) and (2) determine whether the parameters triggering alerts were clinically related to deterioration. Design CDEs are unplanned transfers to the intensive care unit with noninvasive ventilation, tracheal intubation, and/or vasopressor infusion in the 12 hours after transfer. Using one year of data from a large freestanding children's hospital without the pRI, we analyzed CDEs that would have been preceded by pRI alerts. We (1) compared the timing of pRI alerts to time-stamped notes describing changes in patient status and orders reflecting escalations of care and (2) identified score component(s) that caused alerts to trigger and determined whether these were clinically related to CDE etiology. Results Fifty CDEs would have triggered pRI alertsif the pRI had been in use (sensitivity 68%). In 90% of CDEs, the first clinician note reflecting change in patient status and/or the first order reflecting escalation of care preceded the first pRI alert. All of the vital sign and laboratory components of the pRI and 51% of the nursing components were clinically related to the etiology of the CDE. Conclusions Evidence that clinicians were awareof deterioration preceded pRI alerts in most CDEs that generated alerts in the preceding 24 hours.
- Published
- 2018