Large Language Models To Improve the Quality of Care of Cardiology Patients

NCT06935253 | Recruiting

• 1 other identifier: 78695
• Study Type: interventional
• Target: 12
• Locations: 1 country
• Sites: 1

Brief Summary

This study evaluates the impact of large language models (LLMs) versus traditional decision support tools on clinical decision-making in cardiology. General cardiologists will be randomized to manage real patient cases from a cardiovascular genetic cardiomyopathy clinic, with or without AI assistance. Each case will be assessed by two cardiologists, and their responses will be graded by blinded subspecialty experts using a standardized evaluation rubric.

Conditions

Hypertrophic Cardiomyopathy (HCM); Cardiomyopathy; Genetic Disease; Cardiology

Keywords

Large language models; Cardiology

Outcome Measures

Primary Outcomes (1)

• Subspecialist Preference

  The primary outcome is the preference of the subspecialist between answers provided by a) Cardiologist with access to Large Language Model vs. b) Cardiologist without access to Large Language Model.

  Subspecialist evaluation will occur within 1 month of participant completing their assessment

Secondary Outcomes (1)

• Participants perspective on use of Large Language model

  Within one-hour

Study Arms (2)

Large Language Model

ACTIVE COMPARATOR

This group will be given access to a Large Language Model

Other: Large Language Model

Usual resources

NO INTERVENTION

Group will not be given access to a Large Language Model but will be encouraged to use any resources they usually use in their practice besides large language models (UpToDate, Dynamed etc).

Interventions

Large Language Model OTHER

The intervention is a Large Language Model.

Also known as: AMIE (Articulate Medical Intelligence Explorer)

Large Language Model

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Board certified or board eligible Cardiologist.

You may not qualify if:

• Not currently practicing clinically

Sponsors & Collaborators

• Stanford University: lead
• Google LLC.: collaborator

Study Sites (1)

Stanford

Palo Alto, California, 94303, United States

RECRUITING

MeSH Terms

Conditions

Cardiomyopathy, Hypertrophic; Cardiomyopathies; Genetic Diseases, Inborn

Interventions

Large Language Models

Condition Hierarchy (Ancestors)

Heart Diseases; Cardiovascular Diseases; Aortic Stenosis, Subvalvular; Aortic Valve Stenosis; Aortic Valve Disease; Heart Valve Diseases; Congenital, Hereditary, and Neonatal Diseases and Abnormalities

Intervention Hierarchy (Ancestors)

Deep Learning; Machine Learning; Artificial Intelligence; Algorithms; Mathematical Concepts; Neural Networks, Computer

Study Officials

• Euan A Ashley, BSc, MB ChB, DPhil

  Stanford University

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Jack W O'Sullivan, MBBS, DPhil

CONTACT

+16507367878; jackos@stanford.edu

Euan A Ashley, BSc, MB ChB, DPhil

CONTACT

+16507367878; deptmedchair@stanford.edu

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Masking Details: The evaluation of responses will be performed by assessors blinded to participant identity and treatment assignment.
• Purpose: SUPPORTIVE CARE
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Postdoctoral Fellow

Study Record Dates

• First Submitted: April 11, 2025
• First Posted: April 20, 2025
• Study Start: January 10, 2025
• Primary Completion: November 1, 2025
• Study Completion: December 1, 2025
• Last Updated: May 15, 2025

Record last verified: 2025-05

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, ANALYTIC CODE
• Time Frame: The deidentified patient cases and statistical analysis code will be made available within 6 months of study completion.
• Access Criteria: It will be made publicly available and accessible by all.

Locations

US (1)