Plan-Do-Study-Act (PDSA) prompting: a structured approach to prompt engineering in medical education

Large language models are increasingly used in medical education to generate learning resources. However, informal prompt design undermines reproducibility and educational value. This paper introduces 'PDSA Prompting', adapting the Plan-Do-Study-Act quality improvement cycle to systematise prompt engineering. The framework involves a structured, iterative process: defining educational goals, testing prompts, evaluating outputs against predefined criteria, and documenting adaptations. Worked examples demonstrate that this method produces valid assessment materials and feedback. Ultimately, PDSA Prompting aligns artificial intelligence use with established norms of transparency, reproducibility, and accountability.