COVID-19 Mortality Prediction Model

Completed

• Conditions: Mechanical Ventilation; Pneumonia; COVID-19
• Interventions: Device: COViage

• Registration Number: NCT04358510
• Lead Sponsor: Dascena

Brief Summary

The objective of this study is to develop and evaluate an algorithm which accurately predicts mortality in COVID-19, pneumonia and mechanically ventilated ICU patients.

Detailed Description

Retrospective study of 53,001 total ICU patients, including 9,166 patients with pneumonia and 25,895 mechanically ventilated patients, performed on the MIMIC dataset. The NPH patient dataset includes 114 patients positive for SARS-COV-2 by PCR test.

Study Timeline

• Status Verified: 2020-04
• Study Start: 2020-04-01
• Primary Completion: 2020-04-17
• Study Completion: 2020-04-17
• Study First Submitted: 2020-04-20
• Study First Submitted QC: 2020-04-21
• Last Update Submitted: 2020-04-21
• Study First Posted: 2020-04-24
• Last Update Posted: 2020-04-24

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 114

Inclusion Criteria

• Patients aged 18 years or older
• Record of ICU stay

Exclusion Criteria

• Patients aged less than 18 years
• Patients for which there were no records of raw data or no discharge or death dates.

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Arm && Interventions

Group	Intervention	Description
COViage	COViage	Machine learning intervention

Primary Outcome Measures

Name	Time	Method
Mortality outcome in pneumonia ICU patients	Through study completion, an average of 2 months	Deceased or not deceased
Mortality outcome in COVID-19 ICU patients	Through study completion, an average of 2 months	Deceased or not deceased
Mortality outcome in mechanically ventilated ICU patients	Through study completion, an average of 2 months	Deceased or not deceased

Trial Locations

Locations (1)

Dascena; 🇺🇸 Oakland, California, United States