A retrospective feedback analysis of objective structured clinical examination performance of undergraduate medical students.

Abstract

Introduction: Feedback is an essential component of medical education, enhancing the quality of students' knowledge and skills. However, providing effective feedback, particularly in clinical skills assessments like Objective Structured Clinical Examinations [OSCEs], often poses challenges. This study aimed to evaluate the content of OSCE feedback given to undergraduate medical students over five years.

Methods: A retrospective analysis of 1034 anonymised medical students' OSCE performance was conducted, focusing on written feedback. The written feedback data were randomly selected from OSCE sessions, collected from university assessment records and anonymised for ethical considerations. R software was used to identify the most frequently repeated words in the examiners' feedback text, and word cloud charts were created to visualise the responses.

Results: Word clouds generated from the top 200 most frequently used terms provided visual insights into common descriptive words in feedback comments. The most frequently repeated word over five years was "good," indicative of potentially non-specific feedback.

Discussion: The high frequency of non-specific terms like "good" suggests a need for more specific, constructive feedback. However, such generic terms can offer some positive reinforcement, more than they may be needed to foster significant improvement. As previously proposed in the literature, adopting structured feedback forms may facilitate the delivery of more specific, actionable feedback.

Conclusion: This study emphasises the importance of providing specific, actionable feedback in medical education to facilitate meaningful student development. As medical education continues to evolve, refining feedback processes is crucial for effectively guiding students' growth and skill enhancement. Using structured feedback forms can be a beneficial strategy for improving feedback quality.