Abstract
Purpose. This qualitative phenomenological study aimed to explore the integration of artificial intelligence (AI) into hospital administrative workflows. It sought to understand the impact of AI on operational efficiency and identify key barriers and strategies for successful implementation.
Theoretical Framework. The study was guided by a conceptual framework integrating the Consolidated Framework for Implementation Research (CFIR), Socio-Technical Systems (STS) theory, Diffusion of Innovations (DOI) theory, and principles of Human-Centered AI (HCA) and Responsible AI.
Methodology. Data were collected through semi-structured interviews with 18 administrative and clinical personnel, including administrators, IT staff, nurses, and doctors, from two hospital systems in Kansas and Maryland. The data were analyzed using Braun and Clarke's reflexive thematic analysis.
Findings. The analysis revealed that participants viewed AI as a tool for "administrative liberation." It was seen as capable of automating routine tasks, such as documentation and scheduling, thereby allowing more time for patient care. However, significant barriers hindered AI adoption, including outdated technological infrastructure, staff resistance stemming from fear and a lack of understanding, and concerns about data integrity and regulatory compliance. To overcome these barriers, effective strategies included establishing AI governance committees, leveraging early adopters as champions, and providing comprehensive, role-specific training. Participants reported improvements, such as reduced charting time, and anticipated further enhancements in scheduling efficiency and patient flow.
Conclusions and Recommendations. The study concludes that successful integration of AI in hospital administration requires a balanced, human-centric approach that addresses technological, organizational, and ethical challenges. Healthcare leaders should prioritize AI literacy, invest in interoperable infrastructure, and foster multidisciplinary governance. Future research should include longitudinal studies to assess long-term impacts and explore AI adoption in diverse healthcare settings, including rural hospitals.