Early Prediction of Sepsis (ExPRESS)

Early Prediction of Sepsis in Hospitalized Patients Using a Machine Learning Algorithm, a Randomized Clinical Validation Trial.

In this clinical trial a novel Medical Device Software will be validated prospectively. The software incorporates a machine learning algorithm capable of predicting sepsis by using routine clinical variables in adult patients at Intensive Care Units.

Study Overview

• Status: Completed
• Conditions: Sepsis
• Intervention / Treatment: Other: Blinded AlgoDx Sepsis Prediction Algorithm; Device: Unblinded AlgoDx Sepsis Prediction Algorithm

• Study Type: Interventional
• Enrollment (Actual): 320
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Sweden
  • Malmö, Sweden, 20502
    • Intensiv- och perioperativ vård

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 16 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No
• Genders Eligible for Study: All

Description

Inclusion Criteria:

1. Adult patient (age ≥18 years).
2. Patient is admitted to the ICU during the recruitment period of the trial.

Exclusion Criteria:

1. Patient is participating in another interventional clinical trial which, as judged by the investigator, could potentially impact variables used by the sepsis prediction algorithm or has participated in such interventional clinical trial within the last 30 days.
2. Patient is known to be pregnant.
3. Death is deemed imminent and inevitable, at the investigator's discretion.
4. Patient has, due to chronic reduced mental capacity, been assessed by the investigator as incapable of making an informed decision
5. Patient has previously been enrolled in this trial.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Diagnostic
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Triple

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Sham Comparator: Standard of Care; Subjects are monitored for potential development of sepsis according to the local established clinical management guidelines.	Other: Blinded AlgoDx Sepsis Prediction Algorithm; Standard of Care, i.e. no sepsis prediction alert.
Experimental: Standard of Care + AlgoDx Sepsis Prediction Algorithm; Subjects are monitored for potential development of sepsis according to the local established clinical management guidelines, and sepsis prediction algorithm alerts are unblinded to clinical staff.	Device: Unblinded AlgoDx Sepsis Prediction Algorithm; When applicable, a sepsis prediction alert is displayed in the AlgoDx Medical Device Software.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Validate the prognostic accuracy of the algorithm at predicting sepsis.	In order to clinically validate the sepsis prediction performance the following endpoints have been selected: accuracy,; specificity, and; sensitivity of the AlgoDx Sepsis Prediction Algorithm in the SoC group (not possible to assess these in the SoC + Algorithm cohort).	Up to 30 days (ICU hospitalization period)

Collaborators and Investigators

• Sponsor: AlgoDx

Study record dates

Study Major Dates

• Study Start (Actual): December 1, 2020
• Primary Completion (Actual): November 1, 2021
• Study Completion (Actual): November 1, 2021

Study Registration Dates

• First Submitted: September 22, 2020
• First Submitted That Met QC Criteria: September 25, 2020
• First Posted (Actual): September 30, 2020

Study Record Updates

• Last Update Posted (Actual): November 3, 2021
• Last Update Submitted That Met QC Criteria: November 2, 2021
• Last Verified: November 1, 2021

More Information

Terms related to this study

• Keywords: Prediction
• Additional Relevant MeSH Terms: Pathologic Processes; Infections; Systemic Inflammatory Response Syndrome; Inflammation; Sepsis; Toxemia
• Other Study ID Numbers: SEP-SE-02

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: No

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No