AI potential to revolutionise patient involvement in research
A pioneering new approach developed by an NHS Golden Jubilee volunteer could transform how healthcare researchers engage with patients and the public.
This novel application of Artificial Intelligence (AI) large language models offers a virtual and innovative approach to longstanding challenges in Patient and Public Involvement and Engagement (PPIE), combining cutting-edge technology with ethical best practices to enhance diversity, representation, and efficiency.
Traditional PPIE panels — vital for ensuring research reflects the needs of real people — often struggle with recruitment difficulties, geographic barriers, and the time-intensive nature of coordination. These limitations can delay research and compromise the inclusiveness of public input.
The AI system is designed to enhance, not replace, real patient and public voices, by supporting existing PPIE efforts with rapid, diverse input. In parallel, the project partners remain committed to strengthening and diversifying traditional PPIE panels to ensure lived experiences continue to inform research in meaningful and authentic ways.
The creator of the new approach, Andrew Steele, who is also a heart transplant recipient, said: “The new prototype system directly addresses these barriers by generating realistic, AI-simulated PPIE panels.
“Using advanced Large Language Models (LLMs) and UK Census data, the system builds demographically representative virtual panels, enriched with specific lived experiences and health conditions.
“It adheres to UK Standards for Public Involvement and provides structured feedback on research proposals, alongside patient-facing materials such as infographics, quizzes, and video storyboards.”
Key features and potential benefits:
- Accessible communication: Patient-facing outputs scored highly in readability and comprehension metrics, helping improve public understanding of health research.
- Diversity and representation: AI-generated panels could reflect a wide range of backgrounds and lived experiences, matching UK census data and overcoming traditional recruitment gaps.
- Speed and efficiency: Feedback that typically takes weeks can now be generated in minutes, dramatically reducing cost and resource demands.
- High-quality insights: Professional PPIE coordinators rated AI-generated feedback as comparable to that of human panels in terms of relevance and value.
Developed in consultation with PPIE experts and individuals with personal health challenges, the system has already been tested using anonymised cardiology research proposals.
Colin Berry, Director of Research at NHS Golden Jubilee and Professor of Cardiology and Imaging at the University of Glasgow, added: "This AI prototype is not about replacing human input but expanding access to it.
"By potentially offering a hybrid model, we can support faster, more inclusive research processes while still valuing the essential role of lived human experience.
“This innovation represents a promising leap forward in making patient involvement more inclusive, accessible, and embedded across the research lifecycle — offering a powerful new tool for healthcare research in the UK and beyond.”
The project is being developed in collaboration with research teams at the University of Glasgow and the University of Edinburgh and is the first step in exploring the potential use of these models to enhance public voices.
The collaborative team are working to ensure that this approach progresses in an ethical way and that potential biases are addressed from the outset.
