600 Participants NeededMy employer runs this trial

LLM-Generated SIC Summary Emails for Cancer

Recruiting at 1 trial location
AA
CJ
Overseen ByCharlotta J Lindvall, MD, PhD
Age: 18+
Sex: Any
Trial Phase: Academic
Sponsor: Dana-Farber Cancer Institute
No Placebo GroupAll trial participants will receive the active study treatment (no placebo)

What You Need to Know Before You Apply

What is the purpose of this trial?

This trial aims to determine if AI-generated email summaries about serious illness conversations can reduce the number of days cancer patients spend in the hospital. Medical teams receive these summaries to ensure patient care aligns with personal goals. The study includes two groups: one receives AI-generated emails, and the other receives standard care without them. Patients with a solid tumor diagnosis admitted to a specific oncology service may qualify for this trial.

As an unphased trial, this study allows patients to contribute to innovative research that could enhance personalized care.

What prior data suggests that LLM-generated SIC summary emails are safe for patients?

Research has shown that summaries of Serious Illness Conversations (SIC) generated by advanced computer programs are generally safe to use. One study found these summaries to be accurate and well-received by patients, delivered within 16 hours of hospital admission. Another review found that recommendations from these summaries are safe in 93% of cases, indicating that the treatment is usually well-tolerated and poses little risk to patients. While these results are promising, discussing any concerns with a doctor remains important.12345

Why are researchers excited about this trial?

Researchers are excited about this trial because it explores a novel approach to enhancing communication between patients and clinicians in cancer care. Unlike current options that rely on standard serious illness conversations initiated by patients or clinicians, this trial uses an LLM-generated summary email to ensure that these important discussions are aligned with the patient's goals. The email prompts clinicians to review and discuss care preferences, potentially leading to more personalized and patient-centered care. This method aims to bridge communication gaps and ensure that treatment decisions are more closely aligned with what matters most to the patient.

What evidence suggests that LLM-generated SIC summary emails are effective for reducing hospital days in cancer patients?

This trial will compare LLM-generated Serious Illness Conversation (SIC) Summary Emails with standard care. Research has shown that AI tools can quickly create summaries of serious illness conversations, often within 16 hours of a patient's hospital admission. Patients appreciate these summaries, and they help doctors understand patient preferences. AI tools enhance communication, making challenging medical discussions easier and more effective. In studies, reviewers rated AI-generated summaries higher and considered them more complete. Overall, AI in healthcare can assist doctors in making better decisions and reducing their workload.45678

Who Is on the Research Team?

AW

Alexi Wright, MD,MPH

Principal Investigator

Dana-Farber Cancer Institute

Are You a Good Fit for This Trial?

This trial is for people with cancer who are admitted to the hospital and have tumor-related fluid buildup. Specific inclusion or exclusion criteria were not listed, so eligibility may depend on a doctor's assessment.

Inclusion Criteria

Admitted to an inpatient solid tumor medical oncology service at BWH (including beds that are considered to be DFCI beds within BWH)
I am 18 years old or older.
DFCI Patients, defined as patients who have a DFCI medical record number
See 2 more

Exclusion Criteria

I have been admitted to the hospital for planned treatments like chemotherapy.

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

1 week

Intervention

LLM-generated SIC summaries are sent to clinical teams to incorporate patient preferences into care plans

90 days

Follow-up

Participants are monitored for hospital readmissions and days in the hospital post-intervention

90 days

What Are the Treatments Tested in This Trial?

Interventions

  • LLM-generated Serious Illness Conversation (SIC) Summary Email

Trial Overview

The study tests if sending clinical teams an email summary—created by artificial intelligence—about important conversations regarding serious illness can help reduce how many days patients spend in the hospital over three months. Patients are randomly assigned to groups.

How Is the Trial Designed?

2

Treatment groups

Experimental Treatment

Active Control

Group I: Arm 1: Intervention ArmExperimental Treatment1 Intervention
Group II: Arm 2: Control ArmActive Control1 Intervention

Find a Clinic Near You

Who Is Running the Clinical Trial?

Dana-Farber Cancer Institute

Lead Sponsor

Trials
1,128
Recruited
382,000+

Citations

LLM-generated serious illness conversation summaries to ...

LLM-generated SIC summaries can be accurately generated, delivered within 16 hours of hospital admission, are received favorably by patients ...

LLM-generated serious illness conversation summaries to ...

LLM-generated SIC summaries can be accurately generated, delivered within 16 hours of hospital admission, are received favorably by patients ...

Nudges to Increase Serious Illness Conversations

We tested whether pairing nudges with an algorithm that predicts the risk of six-month mortality could increase serious illness conversations between ...

AI Standardized Patient Improves Human Conversations in ...

These results suggest that AI-driven tools can enhance complex interpersonal communication skills, offering scalable, accessible solutions to ...

Simulation-Based Evaluation of a Large Language Model ...

Evidence on LLM impact in oncology workflows remains limited. This study evaluated whether LLM-enabled nCH could generate clinical summaries ...

Better Real-time Information on Documentation of Goals of ...

The goal of this study is to test the accuracy of Large Language Model-generated serious illness communication (SIC) summaries,

Exploring the Feasibility and Acceptability of AI-Mediated ...

Serious illness conversations (SICs) align care with patients' values, goals, and preferences, yet they rarely occur in emergency departments ( ...

8.

pubmed.ncbi.nlm.nih.gov

pubmed.ncbi.nlm.nih.gov/39019351/

Content Analysis of Serious Illness Conversation ...

How serious illness conversations are documented in the electronic health record may impact the content captured.