Unsupervised Machine Learning for Clustering of Septic Patients to Determine Optimal Treatment

• ClinicalTrials.gov Identifier: NCT03752489
• Recruitment Status: Not yet recruiting
• First Posted: November 26, 2018
• Last Update Posted: September 23, 2021
• Sponsor: Dascena
• Information provided by (Responsible Party): Dascena

Study Description

Brief Summary:

The focus of this study will be to conduct a prospective, randomized controlled trial (RCT) at Cape Regional Medical Center (CRMC), Oroville Hospital (OH), and UCSF Medical Center (UCSF) in which a fluid treatment-specific algorithm will be applied to EHR data for the detection of severe sepsis. For patients determined to have a high risk of severe sepsis, the algorithm will generate automated voice, telephone notification to nursing staff at CRMC, OH, and UCSF. The algorithm's performance will be measured by analysis of the primary endpoint, reductions in in-hospital mortality.

Condition or disease	Intervention/treatment	Phase
Sepsis; Severe Sepsis; Septic Shock	Diagnostic Test: Treatment-specific InSight; Diagnostic Test: InSight	Phase 2

Study Design

• Study Type: Interventional (Clinical Trial)
• Estimated Enrollment: 51645 participants
• Allocation: Randomized
• Intervention Model: Parallel Assignment
• Masking: Triple (Participant, Care Provider, Investigator)
• Primary Purpose: Diagnostic
• Official Title: Unsupervised Machine Learning for Clustering of Septic Patients to Determine Optimal Treatment
• Estimated Study Start Date: April 1, 2022
• Estimated Primary Completion Date: March 31, 2024
• Estimated Study Completion Date: March 31, 2024

Arms and Interventions

Arm	Intervention/treatment
Experimental: Fluid treatment-specific algorithm; The experimental arm will involve patients monitored by the fluid treatment-customized version of InSight.	Diagnostic Test: Treatment-specific InSight; The InSight algorithm which draws information from a patient's electronic health record (EHR) to predict the onset of severe sepsis, and in this study will be customized to differentiate between clusters of patients who respond similarly to fluids treatment according to the nature of their disease progression.
Active Comparator: Standard InSight; The control arm will involve patients monitored with the standard, non-treatment specific version of InSight.	Diagnostic Test: InSight; The non-customized InSight algorithm which draws information from a patient's electronic health record (EHR) to predict the onset of severe sepsis.

Outcome Measures

Primary Outcome Measures :

1. In-hospital SIRS-based mortality [ Time Frame: Through study completion, an average of 8 months ]
   Mortality attributed to patients meeting two or more SIRS criteria at some point during their stay

Eligibility Criteria

• Ages Eligible for Study: 18 Years and older (Adult, Older Adult)
• Sexes Eligible for Study: All
• Accepts Healthy Volunteers: Yes

Criteria

Inclusion Criteria:

• All adults above age 18 who are a member of one of the clinical subpopulations studied in this trial are eligible to participate in the study.

Exclusion Criteria:

• Under age 18

Contacts and Locations

Contacts

Contact: Qingqing Mao, PhD; 5108269508; qmao@dascena.com

Sponsors and Collaborators

Dascena

Investigators

• Principal Investigator: Qingqing Mao, PhD; Dascena, Inc.

More Information

Publications:

Calvert J, Mao Q, Hoffman JL, Jay M, Desautels T, Mohamadlou H, Chettipally U, Das R. Using electronic health record collected clinical variables to predict medical intensive care unit mortality. Ann Med Surg (Lond). 2016 Sep 6;11:52-57. eCollection 2016 Nov.

Shimabukuro DW, Barton CW, Feldman MD, Mataraso SJ, Das R. Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay: a randomised clinical trial. BMJ Open Respir Res. 2017 Nov 9;4(1):e000234. doi: 10.1136/bmjresp-2017-000234. eCollection 2017.

Mao Q, Jay M, Hoffman JL, Calvert J, Barton C, Shimabukuro D, Shieh L, Chettipally U, Fletcher G, Kerem Y, Zhou Y, Das R. Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department, general ward and ICU. BMJ Open. 2018 Jan 26;8(1):e017833. doi: 10.1136/bmjopen-2017-017833.

• Responsible Party: Dascena
• ClinicalTrials.gov Identifier: NCT03752489
• Other Study ID Numbers: 19-426246
• First Posted: November 26, 2018
• Last Update Posted: September 23, 2021
• Last Verified: September 2021

• Studies a U.S. FDA-regulated Drug Product: No
• Studies a U.S. FDA-regulated Device Product: No

Keywords provided by Dascena:

Dascena; machine learning; fluid administration; clustering algorithm; mortality; diagnostic

Additional relevant MeSH terms:

Sepsis; Toxemia; Infections; Systemic Inflammatory Response Syndrome; Inflammation; Pathologic Processes