326 - Implementation of the Pediatric eCART early warning score in a tertiary children’s hospital

Poster Number: 326
Publication Number: 326.214

Brett K. Palama, University of Chicago Division of the Biological Sciences The Pritzker School of Medicine, Chicago, IL, United States; Priti Jani, University of Chicago, Comer Children's Hospital, Chicago, IL, United States; Monica Gonzalez, Comer Children's Hospital at University of Chicago Medical Center, Chicago, IL, United States; Anoop Mayampurath, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Joseph A. Reid, AgileMD, Chicago, IL, United States; Dana P. Edelson, University of Chicago Division of the Biological Sciences The Pritzker School of Medicine, Chicago, IL, United States; Allison H. Bartlett, Comer Children's Hospital at University of Chicago Medical Center, Chicago, IL, United States

Background: Early warning scores offer an opportunity to identify hospitalized children at risk for clinical deterioration. Pediatric eCART is an automated model that continuously generates risk scores from electronic health record data and has demonstrated superior performance compared to current pediatric early warning scores (Mayampurath et al, Pediatr Crit Care Med 2020). Pediatric eCART was recently implemented in the acute care wards of our children’s hospital; however, it is unknown whether our implementation approach was successful.

Objective: To analyze the implementation of Pediatric eCART and an associated clinical pathway on the acute care wards of a tertiary children’s hospital. Within six months of go-live, we aimed to (1) reduce the median time to first view of elevated Pediatric eCART scores to 60 minutes, and (2) increase the percentage of time with timely clinical pathway completion (“disposition compliance”) to 55%.

Design/Methods: We implemented Pediatric eCART in a 61-bed acute care unit at a tertiary children’s hospital. Pediatric eCART (AgileMD, San Francisco, CA) was integrated into the electronic health record (Epic Systems, Verona, WI) to generate risk scores with any updates in vital sign or laboratory data. We convened an interdisciplinary stakeholder group to define values for elevated risk scores based on historical local data and develop a nurse-driven clinical pathway to guide response to an elevated score (Figure 1). We completed several improvements after go-live, including priority display of scores within Epic, color-coded display of scores on unit tracker boards, review of performance at unit-level huddles, and continued education on interpretation and response to elevated scores. For each key metric, we generated run charts and reviewed weekly with key stakeholders.

Results: Pediatric eCART was released in our EMR on April 29, 2021. During the study period, the average daily census was 48 patients; a median of 3 patients per day had at least one elevated Pediatric eCART score.  We consistently reduced median time to first view to under 60 minutes by the fifth month following go-live (Figure 2). We are close to meeting our goal for disposition compliance of >55% of hours with elevated Pediatric eCART scores (Figure 3).Conclusion(s): Pediatric eCART was successfully implemented in our general wards, having achieved our goal for median time to first view and approaching our goal for disposition compliance after a process including stakeholder engagement, education, and EHR improvements. Studies on the impact of Pediatric eCART on reducing critical deterioration are ongoing.
Pediatric eCART clinical pathway.Nurse-driven clinical pathway for response to elevated Pediatric eCART score. GCS: Glasgow Coma Scale; PET: Pediatric Early-Response Team; FCP: First Contact Provider. In the setting of an elevated score, the pathway is considered to be complete and up-to-date when checkboxes for clinical concern and disposition have been checked within the preceding 4 hours.
Median time to first view of Pediatric eCART pathway upon first sustained eCART elevation.Each data point represents the weekly median time between first elevation in Pediatric eCART score and entry into the Pediatric eCART module in Epic. Light blue line represents linear trend line; light green shading represents goal of 1 hour; dark green shading represents stretch goal of 30 minutes.