The project aims to adapt the Information Systems Research (ISR) framework to design AI-modified patient-friendly discharge summaries, with the ultimate goal of minimizing the risk of unintended harm to patients from patient-facing generative AI (gen-AI) tools. The project has two specific aims: first, to pilot a User-Centered Design (UCD) approach, following an adapted ISR framework, to gather user specifications for a prototype patient-facing gen-AI discharge instruction tool (Aim 1); and second, to rigorously evaluate this UCD approach using both a user lens (via member checking interviews) and an implementation science lens (using the Consolidated Framework for Implementation Research or CFIR) (Aim 2). This work seeks to establish a validated approach for obtaining user-centered design preferences, particularly incorporating the needs and values of older adults, to model future AI interventions.
Andrew Baumgartner, MD (Department of Family Medicine, Jacobs School of Medicine and Biomedical Sciences)
Connor Wurst, PhD (Department of Industrial Engineering, School of Engineering and Applied Sciences)
Sabrina Casucci, PhD, MBA (Department of Industrial and Systems Engineering, School of Engineering and Applied Sciences)
Kelly Tenzek, PhD (Department of Communication, College of Arts and Sciences)
Ranjit Singh, MB BChir, MBA (Department of Family Medicine, Jacobs School of Medicine and Biomedical Sciences)
Original Funding Source: National Center for Advancing Translational Sciences (NCATS)
Subaward: University at Buffalo Clinical and Translational Science Institute (CTSI) Translational Pilot Studies Program
[03/01/2026 - 02/28/2028]
Active