Large Language Model-Generated Messages to Improve Guideline-Directed Medical Therapy in Heart Failure

Conditions

• Heart Failure

Eligibility Criteria

Sex

ALL

Age

18 Years - 85 Years

Healthy Volunteers

Not accepted

Interventions

• LLM-GDMT Clinical Decision Support Tool — DEVICE
  Software-only, large language model-based clinical decision support tool that reviews structured and unstructured EHR data for adult heart failure patients and generates brief, clinician-facing messages suggesting opportunities to initiate or optimize guideline-directed medical therapy (GDMT) and highlighting relevant safety considerations. Messages are delivered to cardiology providers via Epic InBasket and/or institutional email prior to eligible outpatient visits. The tool is advisory only and cannot place orders or directly change medications; all treatment decisions remain at the discretion of the treating clinician and patient.

Study Details

This study is an investigator-initiated, cluster-randomized implementation trial evaluating a large language model (LLM)-based clinical decision support (CDS) tool designed to improve guideline-directed medical therapy (GDMT) for adult patients with heart failure seen in outpatient cardiology clinics at Mass General Brigham. For eligible heart failure encounters, the CDS tool reviews existing electronic health record (EHR) data, including diagnoses, medications, vital signs, laboratory results, and recent notes, and generates brief, clinician-facing messages suggesting opportunities to initiate or optimize GDMT and highlighting relevant safety considerations. Messages are delivered to cardiology providers via Epic InBasket and/or institutional email prior to scheduled visits. The tool is advisory only and cannot place orders or change medications automatically; all treatment decisions remain at the discretion of the treating clinician and patient. Cardiology providers are assigned at the provider/clinic level to early implementation of the CDS tool versus usual care (no messages) during the initial phase. The primary outcome is GDMT optimization within 30 days of an index visit. Secondary outcomes include feasibility of CDS generation and delivery and a 30-day safety composite (e.g., heart failure hospitalization, acute kidney injury, hyperkalemia, hypotension or bradyarrhythmia plausibly related to GDMT).

Key Dates

Start date

Jun 1, 2026

Status verified

Apr 2026

Primary completion

Jul 1, 2027

Completion

Dec 1, 2027

Study Design

Enrollment

500 participants (estimated)

Allocation

RANDOMIZED

Intervention model

PARALLEL

Primary purpose

HEALTH_SERVICES_RESEARCH

Arms

• Experimental: Early Implementation
  Providers in this arm receive a large language model-based clinical decision support (LLM-GDMT CDS) intervention. For eligible outpatient heart failure encounters, the CDS tool reviews existing EHR data (diagnoses, medications, vitals, labs, recent notes) and generates a brief, clinician-facing message summarizing HF status, suggesting opportunities to initiate or optimize guideline-directed medical therapy (GDMT), and highlighting safety considerations. Messages are delivered via Epic InBasket and/or institutional email in advance of the visit. The tool is advisory only and cannot place orders or directly change medications; all treatment decisions remain at the discretion of the treating clinician and patient.
• No Intervention: Usual Care (Delayed Implementation)
  Providers in this arm continue usual care and do not receive LLM-GDMT CDS messages during the initial evaluation phase. Eligible outpatient heart failure encounters are managed according to routine clinical practice without additional CDS messages. EHR data from these encounters are used to compute GDMT utilization and safety outcomes for comparison with the early-implementation arm. After the initial evaluation phase is complete, the LLM-GDMT CDS tool may be expanded to providers in this arm as part of routine care.

Primary Outcome Measure

Any GDMT optimization within 30 days of index visit [ Time Frame: 30 days ]

Central Contacts

• Jonathan W Cunningham, MD, MPH
  617-732-8534
  [email protected]

Locations (1)

Find similar trials in Boston, MA

Related Studies

• Product Performance Report: Evaluate Long-term Reliability & Performance of Medtronic Marketed Cardiac Therapy Products
  Recruiting · Medtronic · Birmingham, Alabama
• Identification of Genomic Predictors of Adverse Events After Cardiac Surgery
  Recruiting · Brigham and Women's Hospital · Boston, Massachusetts
• Diffusion MRI in Heart Failure
  EARLY_PHASE1 · Recruiting · Massachusetts General Hospital · Charlestown, Massachusetts
• The EMPOWER Trial - The Carillon Mitral Contour System® in Treating Heart Failure With FMR
  Recruiting · Cardiac Dimensions, Inc. · Gilbert, Arizona