Linking Novel Diagnostics With Data-Driven Clinical Decision Support in the Emergency Department

The primary objective of this study is to validate the use of an electronic clinical decision support (CDS) tool, TriageGO with Monocyte Distribution Width (TriageGO-MDW), in the emergency department (ED). TriageGO-MDW is non-device CDS designed to support emergency clinicians (nurses, physicians and advanced practice providers) in performing risk-based assessment and prioritization of patients during their ED visit. This study will follow an effectiveness-implementation hybrid design via the following three aims (phases), to be executed sequentially:

(Aim 1) Validate the TriageGO-MDW algorithm locally using retrospective data at ED study sites.

(Aim 2) Deploy TriageGO-MDW integrated with the electronic medical record (EMR) and perform user assessment.

(Aim 3) Evaluate TriageGO-MDW in steady state with respect to clinical, process, and perceived utility outcomes.

Study Overview

• Status: Recruiting
• Conditions: Inpatient Hospitalization, Intensive Care Unit Admission, Inpatient Mortality, Sepsis and Septic Shock
• Intervention / Treatment: Other: Usual Care; Other: TriageGO-MDW Clinical Decision Support

• Study Type: Observational
• Enrollment (Anticipated): 300000

Contacts and Locations

• Study Contact: Name: Eric Hamrock; Phone Number: 4013420373; Email: eric.hamrock@stocastic.com

Study Locations

• United States
  • Kansas
    • Kansas City, Kansas, United States, 66160
      • Recruiting
      • Kansas University Medical Center
      • Contact: Nima Sarani, MD
  • Missouri
    • Kansas City, Missouri, United States, 64108
      • Recruiting
      • University Health Truman Medical Center
      • Contact: Kevin O'Rourke, MD

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (ADULT, OLDER_ADULT)
• Accepts Healthy Volunteers: Yes
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

All emergency department visits by adult patients during the study period will be included in our analysis.

Description

Inclusion Criteria: Adult patients receiving care at a study site ED

Exclusion Criteria: None

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Pre-Implementation; Usual care will be provided during all ED patient encounters.	Other: Usual Care; Clinical care without decision support provided by TriageGo-MDW
Post-Implementation; TriageGo-MDW CDS will be made available during all ED patient encounters at two points in the ED care continuum: (1) shortly after arrival during initial ED triage (First Triage) and (2) after initial laboratory results have been populated within the EHR. General illness severity estimates will be provided to nurses at ED triage in the form of recommended triage acuity scores (CDS for First Triage). General illness severity estimates along with estimated risk for specific outcomes including sepsis and septic shock will be presented to clinicians after laboratory results have populated (CDS for Early Assessment). TriageGO-MDW risk estimates will be generated by machine learning algorithms using routinely available clinical data as predictor inputs. Nurses and clinicians will receive risk estimates within existing EHR workflows, along with brief and rapidly interpretable explanations of the logic driving each risk estimate.	Other: TriageGO-MDW Clinical Decision Support; TriageGO-MDW is non-device clinical decision support that provides patient-level clinical risk estimates based on clinical data derived from the electronic health record

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Critical Care	Admission to an intensive care unit within 24 hours of ED disposition; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
Critical Care	Admission to an intensive care unit within 24 hours of ED disposition; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)
In-Hospital Mortality	Death during index hospital encounter; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
In-Hospital Mortality	Death during index hospital encounter; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)
Emergent Surgery	procedure in the operating room within 12 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
Emergent Surgery	procedure in the operating room within 12 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)
Sepsis	Prediction performance of machine learning algorithms that underlie TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
Sepsis	Prediction performance of machine learning algorithms that underlie TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)
Septic Shock	Meeting septic shock criteria within 24 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
Septic Shock	Meeting septic shock criteria within 24 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)
Viral Infection	Testing positive for influenza or Covid-19 (SARS-CoV-2) infection within 24 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	baseline (pre-intervention)
Viral Infection	Testing positive for influenza or Covid-19 (SARS-CoV-2) infection within 24 hours of ED arrival; Prediction performance of machine learning algorithms that underly TriageGO-MDW for this outcome will be measured	during post-implementation steady state (approximately 3 months after intervention)

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Critical care triage capture rate	Proportion of patients with critical care admission, emergency surgery or in-hospital mortality identified as high acuity at ED triage	baseline (pre-intervention)
Critical care triage capture rate	Proportion of patients with critical care admission, emergency surgery or in-hospital mortality identified as high acuity at ED triage	during post-implementation steady state (approximately 3 months after intervention)
Hospital admission triage capture rate	Proportion of patients requiring hospital admission identified as moderate or high acuity at ED triage	baseline (pre-intervention)
Hospital admission triage capture rate	Proportion of patients requiring hospital admission identified as moderate or high acuity at ED triage	during post-implementation steady state (approximately 3 months after intervention)
ED patient flow metrics	Intervals between major ED care events, including arrival to disposition, arrival to treatment space, arrival to treatment provider, arrival to intensive care unit transfer, arrival to ED departure will be measured	baseline (pre-intervention)
ED patient flow metrics	Intervals between major ED care events, including arrival to disposition, arrival to treatment space, arrival to treatment provider, arrival to intensive care unit transfer, arrival to ED departure will be measured	during post-implementation steady state (approximately 3 months after intervention)
Sepsis care quality metrics	Standard sepsis care quality metrics including time to diagnosis and treatment and rates of compliance with the Centers for Medicare and Medicaid Services (CMS) Sepsis-1 (SEP-1) Core Measure and its components will be measured	baseline (pre-intervention)
Sepsis care quality metrics	Standard sepsis care quality metrics including time to diagnosis and treatment and rates of compliance with the Centers for Medicare and Medicaid Services (CMS) Sepsis-1 (SEP-1) Core Measure and its components will be measured	during post-implementation steady state (approximately 3 months after intervention)

Collaborators and Investigators

• Sponsor: Stocastic, LLC
• Collaborators: University of Kansas; Truman Medical Center; Beckman Coulter, Inc.
• Investigators: Principal Investigator: Scott Levin, PhD, Stocastic, LLC; Principal Investigator: Jeremiah Hinson, PhD/MD, Stocastic, LLC; Principal Investigator: Nima Sarani, MD, University of Kansas; Principal Investigator: Kevin O'Rourke, MD, Truman Medical Center

Publications and helpful links

General Publications

• Crouser ED, Parrillo JE, Seymour C, Angus DC, Bicking K, Tejidor L, Magari R, Careaga D, Williams J, Closser DR, Samoszuk M, Herren L, Robart E, Chaves F. Improved Early Detection of Sepsis in the ED With a Novel Monocyte Distribution Width Biomarker. Chest. 2017 Sep;152(3):518-526. doi: 10.1016/j.chest.2017.05.039. Epub 2017 Jun 15.
• Levin S, Toerper M, Hamrock E, Hinson JS, Barnes S, Gardner H, Dugas A, Linton B, Kirsch T, Kelen G. Machine-Learning-Based Electronic Triage More Accurately Differentiates Patients With Respect to Clinical Outcomes Compared With the Emergency Severity Index. Ann Emerg Med. 2018 May;71(5):565-574.e2. doi: 10.1016/j.annemergmed.2017.08.005. Epub 2017 Sep 6.
• Dugas AF, Kirsch TD, Toerper M, Korley F, Yenokyan G, France D, Hager D, Levin S. An Electronic Emergency Triage System to Improve Patient Distribution by Critical Outcomes. J Emerg Med. 2016 Jun;50(6):910-8. doi: 10.1016/j.jemermed.2016.02.026. Epub 2016 Apr 25. Erratum In: J Emerg Med. 2016 Aug;51(2):224.

Study record dates

Study Major Dates

• Study Start (ACTUAL): February 1, 2022
• Primary Completion (ANTICIPATED): January 1, 2023
• Study Completion (ANTICIPATED): January 1, 2024

Study Registration Dates

• First Submitted: February 22, 2022
• First Submitted That Met QC Criteria: April 11, 2022
• First Posted (ACTUAL): April 19, 2022

Study Record Updates

• Last Update Posted (ACTUAL): April 19, 2022
• Last Update Submitted That Met QC Criteria: April 11, 2022
• Last Verified: March 1, 2022

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Pathologic Processes; Disease Attributes; Emergencies
• Other Study ID Numbers: 21-STOC-101

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No