Skip to main content
2025 ASBH Annual Conference Main banner
OSU Master of Arts in Bioethics Ad
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Paper Presentation

Health Humanities

Session: AI: End of Life Care

Predicting Death: Multi-Level Ethical Implications of Utilizing Mortality Prediction Algorithms in End-of-Life Care

Friday, October 24, 2025
9:15 AM - 10:15 AM Pacific Time
Location: B112
Jad Brake, PhD – University of British Columbia; Derek Chien, BA – University of Rochester
Abstract: In recent years, as interests in artificial intelligence (AI) in healthcare continue to grow, several AI-powered mortality predictive algorithms (MPAs) have been developed to predict a patient’s life expectancy, with the hope that algorithmic outputs will help providers to better time the "therapeutic window" to initiate end-of-life conversations to enhance goal-concordant care.
Despite robust discussions on ethical implications of integrating AI into various areas of medicine, relatively little attention has been placed on the ethics of utilizing AI-powered MPAs in the sensitive and vulnerable context of the final stages of a person’s life. Informed by an environmental scan, literature review, and qualitative research study conducted in xxxx regarding clinicians’ perspectives on ethical considerations of using MPAs, this presentation discusses how MPAs raise familiar but also new multi-level ethical issues regarding the use of predictive analytics in clinical medicine. In particular, this presentation explores 1) individual-level concerns around a particular algorithm’s accuracy and potential impact of MPA implementation on individuals’ (e.g., patients, families, clinicians) therapeutic experiences and authority in consenting or disputing algorithmic use and predictions; 2) professional-level concerns around how AI-assisted prognostication may affect clinical workflow, therapeutic relationships, and the ways end-of-life information is presented and communicated; 3) institutional-level considerations around how to manage accountability and disputes, especially since MPAs may pre-determine treatment or service eligibility and affect workflows; 4) societal-level considerations of how increasing promotion and implementation of automated and quantified predictions may affect the concepts of and approaches to hope, acceptance, and uncertainties in end-of-life care.

Keywords: Artificial Intelligence, End-of-Life Care

Learning Objectives:

After participating in this conference, attendees should be able to:
  • Describe how mortality prediction algorithms can be used in clinical care delivery
  • Understand multiple layers of ethical concerns around using AI-powered mortality prediction algorithms.
  • Understand the necessity and practicality of integrating ethical considerations and principles into implementing AI in end-of-life care.