Canada Spends Logo

Explore our Budget 2026 analysis

Grant: DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)

McGill University

$120,000.00

Artificial intelligence (AI)-generated electronic health record summaries to promote high-quality communication and care in serious illness: development, validation, and initial feasibility

FY 2025-26

Summary
Palliative care helps people with serious illnesses, such as advanced cancer, live as well as possible by expertly managing symptoms and building supportive relationships. Yet, the time and communication needed for high-quality palliative care are threatened by practice inefficiencies. Prior to patient visits, physicians must review their complex electronic health records (EHRs). Vital information is scattered across clinical notes, radiology and lab reports, and prescriptions; patients with serious illnesses interact frequently with diverse healthcare providers, leading to a substantial volume of difficult-to-synthesize health records. As a result, preparing for visits is often difficult and time-consuming, which can reduce the face-to-face time available for patients, risk missed information, and threaten the sustainability of an already overstretched palliative care workforce. Recent advances in artificial intelligence (AI) offer a potential solution. Large language models (like ChatGPT) can rapidly process and summarize vast amounts of clinical text. Used to streamline record review, AI-generated patient EHR summaries could reduce time pressure, enrich communication during visits, and support higher quality, relationship-centered care for patients with serious illness. My PhD thesis focuses on developing and evaluating AI-generated EHR summaries informed by the needs of palliative care clinicians and their patients at a large Canadian cancer centre. My project will be guided by an end-user advisory committee. This project will involve: (1) refining a process for generating and assessing summaries informed by real-world patient-physician needs, (2) validating the quality of AI-generated summaries against human-authored summaries using blinded expert ratings, and (3) assessing the feasibility of implementing these summaries and preliminary outcomes regarding visit preparation and communication quality. Findings will inform the design of a clinical trial.
Program
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
560224
Name
Durieux Brigitte
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Health Services and Policy Research
Theme2
Health systems / services
Prcname2
DT2 (Health services and population health research) CGRS Doctoral
Projecttitle
Artificial intelligence (AI)-generated electronic health record summaries to promote high-quality communication and care in serious illness: development, validation, and initial feasibility
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Keyworddelim
Advanced Cancer; Artificial Intelligence; Doctor-Patient Communication; Doctor-Patient Relationships; Feasibility Trial; Focus Groups; Large Language Models; Palliative Care; Patient Priorities; Serious Illness
Abstract
Palliative care helps people with serious illnesses, such as advanced cancer, live as well as possible by expertly managing symptoms and building supportive relationships. Yet, the time and communication needed for high-quality palliative care are threatened by practice inefficiencies. Prior to patient visits, physicians must review their complex electronic health records (EHRs). Vital information is scattered across clinical notes, radiology and lab reports, and prescriptions; patients with serious illnesses interact frequently with diverse healthcare providers, leading to a substantial volume of difficult-to-synthesize health records. As a result, preparing for visits is often difficult and time-consuming, which can reduce the face-to-face time available for patients, risk missed information, and threaten the sustainability of an already overstretched palliative care workforce. Recent advances in artificial intelligence (AI) offer a potential solution. Large language models (like ChatGPT) can rapidly process and summarize vast amounts of clinical text. Used to streamline record review, AI-generated patient EHR summaries could reduce time pressure, enrich communication during visits, and support higher quality, relationship-centered care for patients with serious illness. My PhD thesis focuses on developing and evaluating AI-generated EHR summaries informed by the needs of palliative care clinicians and their patients at a large Canadian cancer centre. My project will be guided by an end-user advisory committee. This project will involve: (1) refining a process for generating and assessing summaries informed by real-world patient-physician needs, (2) validating the quality of AI-generated summaries against human-authored summaries using blinded expert ratings, and (3) assessing the feasibility of implementing these summaries and preliminary outcomes regarding visit preparation and communication quality. Findings will inform the design of a clinical trial.
Competitiondate
202511
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Durieux, Brigitte
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~
McGill University - $120,000 Grant (2025-26) | Canada Spends