Connected Health Approaches to Improve Serious Illness Communication Among Patients with Incurable Cancer
- Augmented & Artificial Intelligence,
- Behavior Change
This project developed & piloted “Conversation Connect”, a machine learning model that identifies patients with incurable cancer who may benefit from a timely Serious Illness Conversation (SIC) with their clinician. Findings will help clinicians better identify palliative care needs & increase rates of documented SICs.
Many patients with incurable cancer undergo treatment and acute care utilization that is discordant with their preferences, particularly near the end of life. Early conversations about prognosis and treatment preferences can improve prognostic awareness between cancer patients and their oncologists, increase goal-concordant care, and decrease utilization near the end of life.
This project developed and piloted “Conversation Connect”, a machine learning model based on structured electronic health record data to identify patients who may benefit from a timely Serious Illness Conversation (SIC) with their clinician. This project also explores how trends in out-of-clinic patient-reported symptoms and step counts can prompt serious illness conversations.
Providing clinicians with patient-reported symptoms and remotely-monitored functional status will help clinicians better identify palliative care needs and increase rates of documented SICs.
Center for Health Incentives and Behavioral Economics (CHIBE)
Project Leads
-
Ravi Parikh
MD, MPP, FACP
Associate Director, PC3I & Director, Program in Augmented and Artificial Intelligence, PC3I
-
Mitesh Patel
MD, MBA
Vice President of Clinical Transformation & National Lead for Behavioral Insights, Ascension Health
Project Team
-
Justin Bekelman
-
Peter Gabriel
-
Christopher Manz
-
Nina O'Connor
-
Larry Shulman
-
Samuel Takvorian
