Poster (Times are in PDT)

Board 26 - Leveraging Data Analytics to Redesign the Electronic Health Record and Improve Pediatric Sepsis Care

3:30 PM–4:30 PM May 20, 2020 (Conference Time: US - Pacific)

3:30 PM–4:30 PM May 20, 2020


Abstract: This project evaluates the scalable, phased-implementation approach to improving early identification and treatment of pediatric sepsis through the application of an enhanced electronic infrastructure, best-practice workflow implementation, and documentation standardization. The institution examines alert statistical performance analysis to assess changes and improve design. Post-implementation time-to-intervention metrics and patient outcomes are discussed.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Participants will learn strategies the institution implemented to improve pediatric sepsis identification, treatment, and patient outcomes through clinical-decision-support to impact practice. Presenters will explain the modified screening criteria developed from evidence-based practice, retrospective case review, and pre-implementation statistical alert analysis to provide predictive performance of the alert screening. Presenters will discuss the institution’s customizations to optimize the enterprise-wide electronic surveillance system, as well as ordering and documentation workflow to guide practice, standardize care delivery, and ensure compliance with pediatric sepsis treatment guidelines. New knowledge and skills include, but are not limited to, applied timed-visual cues in the electronic health record, alert acknowledgement documentation, alert lockout behavior to combat alert fatigue, implementation of the bedside huddle for clinical discernment, as well as education and implementation strategies. Data on rates of patients identified as at-risk for sepsis pre and post-implementation, turnaround time from identification to clinical huddle, turnaround time from identification to treatment, and patient outcomes are examined. Presenters will also assess alert modifications resulting in increased sensitivity, and an increased negative predictive value.


Rachelle Torres (Presenter)
NYU Langone Health

Elizabeth Haines, NYU Langone Health
Allan Michael Flores, NYU Langone Health
Edwin Pineda, NYU Langone Health
Vanesa Flaviano, NYU Langone Health
Mario Legaspi, NYU Langone Health