Abstract
This paper explored students' use of Generative Artificial Intelligence (GenAI) for Objective Structured Clinical Examinations (OSCE) preparation using a retrospective cohort study conducted over two years (2023–2024) across Australia and Malaysia. Results analysed OSCE grades and written self-reflections from students. The Unified Theory of Acceptance and Use of Technology (UTAUT) model was applied as a lens to interpret qualitative data using summative content analysis. Of 997 students, 163 (16.3%) stated they did use GenAI to prepare for OSCEs. Across both campuses, there was no significant difference in mean OSCE grades between GenAI users (79.9%, SD = 15.3) and non-GenAI users (79.6%, SD = 17.4; p = .6); however, non-GenAI users performed significantly better (p < .05) in four of the seven communication rubric criteria. Themes around mistrust or perceived inaccuracy of GenAI data for clinical application deterred use in non-users. From our results, the use of GenAI did not demonstrate additional benefits for overall OSCE preparation, suggesting the potential need for more pedagogically aligned applications of GenAI tools to maximise their utility for clinical assessment preparation.
Keywords
Generative Artificial Intelligence, Objective Structured Clinical Examinations, The Unified Theory of Acceptance and Use of Technology.