Development of a Framework and Checklist to Guide the Translation of AI Systems for Clinical Care

Implementing AI systems in a clinical setting requires adequate consideration of the social aspects of its application alongside the technical ones. It is essential to understand the context in which the technology will be used, how it will work with existing workflows without disruption, and how it will be accepted by the people who will have to use it. Despite the numerous proof-of-concept publications in this field, the lack of robust frameworks for supporting these tools' design, development, and management is one of the main barriers to their adoption in healthcare. This study employs primary research using a mixed methods approach to investigate the factors influencing the implementation and effectiveness of the Epic Sepsis System at UI Health and the experiences of clinicians with the same tool. It combines this with the results of a literature review to create a framework. A Delphi study was conducted to expand the framework into a checklist, which was validated through interviews with purposively selected members of the UI Health Diabetes Tool development team. Findings from this study show that the successful implementation and integration of AI in a clinical setting depends on a balanced focus on the clinical settings' technical and social dynamics. Also identified were critical gaps in existing frameworks primarily focused on technical specifications or ethics, neglecting the comprehensive sociotechnical dynamics essential for developing and implementing AI systems in clinical environments. Clinical AI Sociotechnical Framework (CASoF) addresses this gap by providing a framework and a structured checklist that guides the planning, design, development, and implementation stages of AI systems in clinical settings. The checklist emphasizes the importance of considering the value proposition, data integrity, human-AI interaction, technical architecture, organizational culture, and ongoing support and monitoring, ensuring that AI tools are technologically sound, practically viable, and socially adaptable within clinical settings. CASoF represents a step forward in bridging this divide, offering a holistic approach to AI deployment that is mindful of the complexities of healthcare systems.