214 - Bridging the AI Gap in Healthcare: An Innovative Approach to Educating Medical Professionals on Generative AI

Publication Number: 214.4444

Andrew Tran, University of Arkansas for Medical Sciences College of Medicine, Little Rock, AR, United States; Vasilii Khammad, University of Arkansas for Medical Sciences, North Little Rock, AR, United States; Jonathan Chastain, University of Arkansas for Medical Sciences College of Medicine, Little Rock, AR, United States; Joshua Ramsay, Arkansas Children's Hospital, Little Rock, AR, United States; David Becton, Arkansas Children's Hospital, LITTLE ROCK, AR, United States; Feliciano Yu, University of Arkansas for Medical Sciences, Little Rock, AR, United States; Joana Mack, Arkansas Children's Hospital, Little Rock, AR, United States

Background: As artificial intelligence (AI) continues to transform healthcare, there is a growing need for medical professionals to understand and effectively utilize these technologies. However, a significant knowledge gap exists between technological advancements and healthcare workers' familiarity with AI tools.

Objective: Our study aimed to assess the impact of innovative, tailored educational interventions on healthcare professionals' perceptions, understanding, and willingness to adopt generative AI technologies.

Design/Methods: We conducted a series of educational sessions, including grand rounds and conference presentations, for diverse healthcare audiences at Arkansas Children's Hospital and the Arkansas Chapter of the American Academy of Pediatrics. The sessions employed creative elements such as short-form videos and animated hand drawings to make the content relatable. Each presentation included specialty-specific use cases and interactive demonstrations. Pre and post-intervention surveys measured comfort levels, understanding, and perceived usefulness of AI.

Results: A total of 100 participants completed the pre-survey, and 90 completed the post-survey. Comfort with using generative AI increased by 36.18% (95% CI: 2.03-2.49), from a mean of 2.26 to 3.08 on a 5-point scale. Understanding of generative AI improved by 54.94% (95% CI: 1.96-2.40), from a mean of 2.18 to 3.38. Willingness to use AI in healthcare domains also increased: Scholarly Work by 21.11% (from 60.00% to 81.11%, 95% CI: 0.50-0.70), Clinical Work by 20.22% (from 52.00% to 72.22%, 95% CI: 0.42-0.62), and Administrative Tasks by 12.44% (from 82.00% to 94.44%, 95% CI: 0.74-0.90). Skepticism (defined as Disagree/Strongly Disagree responses) decreased, with a reduction in Clinical Work responses from 16 to 8, and a drop in Scholarly Work from 12 to 1. Agreement that healthcare organizations should implement formal systems for safe AI use increased by 5.17% (95% CI: 4.37-4.70), from 4.54 to 4.77.

Conclusion(s): This study demonstrates that creatively designed educational sessions can significantly improve healthcare professionals' comfort with and understanding of AI technologies. The approach's effectiveness across specialties suggests a scalable model for AI literacy in healthcare. These findings have important implications for graduate medical education programs, highlighting the need for innovative AI curricula to prepare healthcare professionals for an AI-integrated landscape. Future steps include advanced sessions to increase proficiency with AI tools, as well as applying this educational model to other novel technologies in healthcare.

Attitudes Towards Generative AI Usage in Academic, Clinical, and Administrative Settings: A Pre- and Post-Survey Comparison
This figure illustrates changes in attitudes towards the potential use of generative AI for scholarly work, clinical tasks, and administrative responsibilities, comparing responses before and after the educational experience. Notable increases in 'Agree' and 'Strongly Agree' ratings suggest a growing openness to leveraging AI across these domains, particularly in administrative tasks.