Prediction of Patient Deterioration Using Machine Learning
1 other identifier
observational
526
1 country
2
Brief Summary
This is a retrospective observational study drawing on data from the Brigham and Women's Home Hospital database. Sociodemographic and clinic data from a training cohort were used to train a machine learning algorithm to predict patient deterioration throughout a patient's admission. This algorithm was then validated in a validation cohort.
Trial Health
Trial Health Score
Automated assessment based on enrollment pace, timeline, and geographic reach
participants targeted
Target at P75+ for all trials
Started Mar 2021
Longer than P75 for all trials
2 active sites
Health score is calculated from publicly available data and should be used for screening purposes only.
Trial Relationships
Click on a node to explore related trials.
Study Timeline
Key milestones and dates
Study Start
First participant enrolled
March 20, 2021
CompletedFirst Submitted
Initial submission to the registry
April 14, 2021
CompletedFirst Posted
Study publicly available on registry
September 16, 2021
CompletedPrimary Completion
Last participant's last visit for primary outcome
March 20, 2025
CompletedStudy Completion
Last participant's last visit for all outcomes
February 16, 2026
CompletedMarch 17, 2026
March 1, 2026
4 years
April 14, 2021
March 16, 2026
Conditions
Outcome Measures
Primary Outcomes (1)
Alarm burden
The number of alarms fired per patient per hour
From admission to discharge, measured in hours, on average 5 days
Secondary Outcomes (5)
Sensitivity for recognition of a safety composite
From admission to discharge, on average 5 days
Specificity for recognition of a safety composite
From admission to discharge, on average 5 days
Positive predictive value for recognition of a safety composite
From admission to discharge, on average 5 days
Negative predictive value for recognition of a safety composite
From admission to discharge, on average 5 days
Rate of alarms with clinical utility
From admission to discharge, on average 5 days
Study Arms (2)
Training
A subset of patients that are used to train the machine learning algorithm.
Validation
A subset of patients that are "held back" and used to validate the algorithm's accuracy.
Interventions
We will retrospectively compare the alarms produced from traditional vital sign alarms (thresholds set by clinicians) versus the BioVitals Index vs the National Early Warning Score 2
Eligibility Criteria
Subjects admitted at Brigham and Women's Hospital and Brigham and Women's Faulkner Hospital who meet primary diagnosis, age, and geographic residence requirements and are enrolled in home hospital.
You may qualify if:
- Cared for in the Brigham and Women's Home Hospital study
You may not qualify if:
- Incomplete continuous monitoring data
Contact the study team to confirm eligibility.
Sponsors & Collaborators
- Brigham and Women's Hospitallead
- Biofourmis Inc.collaborator
Study Sites (2)
Brigham and Women's Hospital
Boston, Massachusetts, 02115, United States
Brigham and Women's Faulkner Hospital
Boston, Massachusetts, 02130, United States
MeSH Terms
Conditions
Condition Hierarchy (Ancestors)
Study Officials
- PRINCIPAL INVESTIGATOR
David Levine, MD MPH MA
Associate Physician
Study Design
- Study Type
- observational
- Observational Model
- COHORT
- Time Perspective
- RETROSPECTIVE
- Sponsor Type
- OTHER
- Responsible Party
- PRINCIPAL INVESTIGATOR
- PI Title
- Attending Physician
Study Record Dates
First Submitted
April 14, 2021
First Posted
September 16, 2021
Study Start
March 20, 2021
Primary Completion
March 20, 2025
Study Completion
February 16, 2026
Last Updated
March 17, 2026
Record last verified: 2026-03
Data Sharing
- IPD Sharing
- Will not share