Effect of a Sepsis Prediction Algorithm on Clinical Outcomes

Not Applicable

Completed

• Conditions: Severe Sepsis

• Registration Number: NCT03960203
• Lead Sponsor: Dascena

Brief Summary

In this clinical outcomes analysis, the effect of a machine learning algorithm for severe sepsis prediction on in-hospital mortality, hospital length of stay, and 30-day readmission was evaluated.

Detailed Description

Materials and Methods: Clinical outcomes evaluation performed on a multiyear, multicenter clinical data set of real-world data containing 75,147 patient encounters from nine hospitals. Mortality, hospital length of stay, and 30-day readmission analysis performed for 17,758 adult patients who met two or more Systemic Inflammatory Response Syndrome (SIRS) criteria at any point during their stay.

Study Timeline

• Study Start: 2017-01
• Primary Completion: 2018-06
• Study Completion: 2018-06
• Status Verified: 2019-05
• Study First Submitted: 2019-05-17
• Study First Submitted QC: 2019-05-21
• Study First Posted: 2019-05-22
• Last Update Submitted: 2019-05-22
• Last Update Posted: 2019-05-24

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 75147

Inclusion Criteria

• All patients over the age of 18 presenting to the emergency department or admitted to an inpatient unit at the participating facilities were automatically included for clinical outcomes analysis

Exclusion Criteria

• Patients under the age of 18

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
In-hospital mortality	1 year	Rate of in-hospital mortality based on SIRS criteria

Secondary Outcome Measures

Name	Time	Method
30-day readmissions	1 year	Rate of patient readmissions within 30 days
Hospital length of stay	1 year	Duration of hospital length of stay in days based on SIRS criteria