Utility of large language models for creating clinical assessment items

To compare student performance, examiner perceptions and cost of GPT-assisted (generative pretrained transformer-assisted) clinical and professional skills assessment (CPSAs) items against items created using standard methods.