MedPath

Predicting Favorable Outcomes in Hospitalized Covid-19 Patients

Completed
Conditions
COVID
Corona Virus Infection
Adverse Event
Interventions
Other: EPIC risk score display
Registration Number
NCT04570488
Lead Sponsor
NYU Langone Health
Brief Summary

Testing use of predictive analytics to predict which COVID-19+ patients are at low risk for an adverse event (ICU transfer, intubation, mortality, hospice discharge, re-presentation to the ED, oxygen requirements exceeding nasal cannula at 6L/Min) in the next 96 hours

Detailed Description

To assess if display of low risk of adverse event in EPIC can safely reduce length of stay and plan for discharge.

Recruitment & Eligibility

Status
COMPLETED
Sex
All
Target Recruitment
1415
Inclusion Criteria

Adult hospitalized COVID19+ patients predicted to have no adverse event at 96 events with a threshold at 90% PPV, with at least one low risk during their admission who are discharged alive and have not been in the ICU

Exclusion Criteria

Age < 18 years not hospitalized for COVID19+.

Study & Design

Study Type
OBSERVATIONAL
Study Design
Not specified
Arm && Interventions
GroupInterventionDescription
Quality improvement - DisplayEPIC risk score displayDisplay of risk score/ colored flag in Epic patient list column; will be viewable to all frontline workers
Primary Outcome Measures
NameTimeMethod
Reduction in days from first low-risk score to discharge96 Hours

Reduction in days from first low-risk score to discharge

Secondary Outcome Measures
NameTimeMethod
Reduction in length of stay (LOS)96 hours

Reduction in LOS for green patients that have not been in the ICU

Reduction in GTD vs. LOS96 hours

Reduction in GTD vs. LOS for all green patients discharged alive vs all patients discharged alive

No change in 30 day re-ED presentation or hospital admission rate for cohort96 hours

No change in 30 day re-ED presentation or hospital admission rate for cohort

Trial Locations

Locations (1)

NYU Langone Health

🇺🇸

New York, New York, United States

© Copyright 2025. All Rights Reserved by MedPath