Using Arts-Based Methods to Involve People Living in Tower Hamlets With Multiple Long-Term Conditions in the...

Area of Science:

• Healthcare research
• Artificial intelligence (AI)
• Public and patient involvement and engagement (PPIE)

Background:

• Traditional public involvement methods in healthcare AI research often exclude minoritized groups.
• There is a growing need for inclusive approaches to engage diverse populations in AI development.

Purpose of the Study:

• To explore the use of art-based methods to involve under-represented communities in healthcare AI research.
• To gather perspectives from individuals with lived experience of multiple long-term conditions (MLTCs) and polypharmacy.

Main Methods:

• A collaboration with a community organization in East London involved 22 PPIE contributors from under-represented backgrounds.
• Six workshops utilized visual arts (ceramics, collage, body mapping, AI-generated images) to facilitate input on AI development.
• Contributors had lived experience of MLTCs and represented diverse ages, genders, socio-demographics, and multilingual abilities.

Main Results:

• Arts-based methods facilitated communication and enabled PPIE contributors to define research priorities, such as the impact of appointment frequency on MLTCs.
• Contributors highlighted key considerations for AI model development, including the use of local housing data and addressing bias.
• Visual arts made abstract AI concepts more tangible, enhancing AI literacy and engagement.

Conclusions:

• Arts-based methods are effective in making research participation more accessible for under-represented communities in healthcare AI.
• Inclusive participatory approaches are crucial as AI in healthcare research continues to expand.
• The study successfully generated research priorities and informed AI development through novel engagement strategies.