LLM SIC Summary Email for Cancer
(BRIDGE-SIC Trial)
What You Need to Know Before You Apply
What is the purpose of this trial?
This trial evaluates how effectively artificial intelligence (AI) can create summaries of serious illness conversations for cancer patients. The goal is to determine if these AI-generated summaries are accurate, easy to use, and helpful in cancer care. Participants will either receive the AI-generated summary or continue with their usual cancer care, and they can share their thoughts in an interview. The trial seeks cancer patients admitted to specific hospital services, who have a higher risk of not surviving, and who are admitted at certain times of the week. As an unphased trial, this study offers patients the opportunity to contribute to innovative research that could enhance communication in cancer care.
Will I have to stop taking my current medications?
The trial information does not specify whether you need to stop taking your current medications. It's best to discuss this with the trial coordinators or your doctor.
What prior data suggests that this method is safe for generating SIC summaries?
Research has shown that large language models (LLMs) are already used in healthcare, particularly to summarize patient preferences in care. These AI tools enhance communication between patients and doctors by simplifying complex information.
Other studies have used LLMs to assist with cancer treatment decisions, demonstrating an accuracy of about 76.2%, which indicates they can reliably summarize important information.
Regarding safety, AI tools like LLMs do not pose direct health risks as a new drug might. As software tools, the main concern is their accuracy in summarizing information. No reports of harmful effects exist since they do not interact with the body. The focus remains on ensuring the information they provide is correct and helpful.
Overall, LLMs are considered safe because they are a type of technology, not a medical treatment. It is important to ensure they support, rather than replace, human judgment in healthcare.12345Why are researchers excited about this trial?
Researchers are excited about this trial because it explores a new way of communicating complex cancer care information using AI-generated summaries. Unlike traditional methods that rely solely on healthcare professionals to explain treatment options, this approach uses language model-generated emails to summarize and clarify information for patients. This could potentially make understanding treatment plans easier and more accessible, offering a personalized way to digest complex medical information. By improving how patients perceive and accept their care details, this trial might pave the way for more informed and empowered patient decision-making in oncology.
What evidence suggests that this method is effective for cancer communication?
Research has shown that large language models (LLMs) can assist in cancer decision-making with approximately 76.2% accuracy. This indicates their reliability in summarizing important health information for patients. In this trial, participants in Group A will receive LLM-generated SIC summaries, potentially enhancing communication between patients and doctors. Meanwhile, Group B will receive usual care without LLM-generated summaries. Past projects have used LLMs to improve discussions about advanced cancer care. These findings suggest that LLMs could effectively create clear and helpful summaries for discussing serious illnesses.16789
Who Is on the Research Team?
Christopher Manz, MD
Principal Investigator
Dana-Farber Cancer Institute
Are You a Good Fit for This Trial?
This trial is for individuals with cancer, particularly those facing end-of-life decisions. Participants should be able to receive and review email communications as part of the study. The specific inclusion and exclusion criteria are not detailed here.Inclusion Criteria
Exclusion Criteria
Timeline for a Trial Participant
Screening
Participants are screened for eligibility to participate in the trial
Intervention
Participants are randomized to intervention or control arms. Intervention arm receives LLM-generated SIC summaries.
Follow-up
Participants are monitored for new documentation of SIC and other outcomes during hospitalization.
Extended Follow-up
Participants' hospitalizations, ICU admissions, and other health outcomes are tracked over 180 days.
What Are the Treatments Tested in This Trial?
Interventions
- LLM SIC Summary
Trial Overview
The study is testing the use of Large Language Models (LLMs) to generate summaries about serious illness communication (SIC). It aims to assess how accurate these AI-generated summaries are, if they can be feasibly delivered via email, and what clinicians and patients think about them.
How Is the Trial Designed?
2
Treatment groups
Experimental Treatment
Active Control
45 participants will be randomized. Participants will have the option of completing a one-time in-person interview with study staff on their perceptions of SIC and the acceptability of LLM-generated summaries of SIC.
15 participants will be randomized and will receive standard oncology care. Participants will have the option of completing a one-time in-person interview with study staff on their perceptions of SIC and the acceptability of LLM-generated summaries of SIC.
Find a Clinic Near You
Who Is Running the Clinical Trial?
Dana-Farber Cancer Institute
Lead Sponsor
Citations
Large language model integrations in cancer decision-making
The results of the meta-analysis indicated that the overall average accuracy rating for LLM-assisted cancer decision making was 76.2% (95% CI: ...
LLM SIC Summary Email for Cancer (BRIDGE-SIC Trial)
The goal of this study is to test the accuracy of Large Language Model-generated serious illness communication (SIC) summaries, the feasibility of ...
3.
acsjournals.onlinelibrary.wiley.com
acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncr.70050Artificial intelligence across the cancer care continuum - Riaz
This review highlights the expanding role of artificial intelligence across the cancer care continuum, from risk prediction to end-of-life ...
4.
researchgate.net
researchgate.net/publication/391461563_AI_Standardized_Patient_Improves_Human_Conversations_in_Advanced_Cancer_CareAI Standardized Patient Improves Human Conversations in ...
SOPHIE combines large language models (LLMs), a lifelike virtual avatar, and automated, personalized feedback based on clinical literature to ...
Develop an end-to-end LLM-powered application tailored for ...
Our application automates the generation of LLM prompts from the Cancer patient dataset, enabling clinicians to access relevant information quickly and ...
6.
qa-financial.com
qa-financial.com/ai-summaries-of-patient-preferences-set-to-transform-qa-in-healthcare/AI summaries of patient preferences set to transform QA in ...
A new AI pilot is redefining QA by using LLMs to summarise patient care preferences…
A web-based, LLM-powered AI symptom summarization ...
A web-based, LLM-powered AI symptom summarization tool (ASST) for monitoring of breast cancer treatment toxicity. download. Background: Traditional methods for ...
8.
acsjournals.onlinelibrary.wiley.com
acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncr.35307Uses and limitations of artificial intelligence for oncology - Kolla
The authors review the current potential and limitations of predictive AI for cancer diagnosis and prognostication as well as of generative AI.
AI Standardized Patient Improves Human Conversations in ...
Low-quality SIC has been associated with poor patient and family prognostic understanding [5] , perceived lack of emotional support [6] , lower ...
Unbiased Results
We believe in providing patients with all the options.
Your Data Stays Your Data
We only share your information with the clinical trials you're trying to access.
Verified Trials Only
All of our trials are run by licensed doctors, researchers, and healthcare companies.