Exploring prompt literacy in allied health professions education: an explanatory sequential mixed-methods study of student and educator perspectives

Background Generative artificial intelligence (GenAI) is increasingly used in health professions education, but students’ experiences of prompting and their ability to effectively or critically evaluate AI-generated outputs remain underexplored. This study examined prompt-related experiences, challenges, and educational needs among allied health students and academic staff. Methods An explanatory sequential mixed-methods study was conducted at a public university in Ghana. Quantitative data were obtained from 417 undergraduate students. Of these, 261 provided responses to an open-ended question concerning challenges associated with AI use. Out of the 261 open-ended responses, 205 responses contained relevant descriptions of challenges that were included in the content analysis. Semi-structured interviews were conducted with 18 third- and fourth-year students and eight academic staff across six allied health disciplines. Quantitative data were analysed descriptively, open-ended responses using content analysis, and interviews using reflexive thematic analysis. Findings relevant to prompt literacy were integrated interpretively. Results Most surveyed students (76.3%, n = 318) reported no prior training in responsible AI use. Integration produced four explanatory categories: access without consistent prompting capability; prompting as purposeful dialogue for personalised learning; prompt quality, prior knowledge, and verification as interdependent safeguards; and prompt literacy as a shared curricular and institutional responsibility. Participants described using prompts to simplify difficult concepts, tailor explanations, and generate practice questions, alongside difficulties formulating prompts and verifying outputs. Their accounts suggested a prompt-verification paradox, whereby students may seek AI support when their understanding is limited, while simultaneously being less able to recognise inaccurate information. Conclusion Participants’ accounts suggest that prompt literacy involves more than constructing effective prompts; it also includes purposeful task framing, iterative dialogue, disciplinary grounding, verification, ethical awareness, and human judgement. Allied health curricula should progressively develop these capabilities within broader AI literacy while providing educator and institutional support for responsible GenAI use.