How Supervisors Work With Video of Clinical Practice to Influence Learning Conversations

Feedback in clinical practice is often sub-optimal due to workload, time constraints, and interpersonal dynamics. Recent work has explored the concept of 'learning conversations' as a way to bring the learner into the dialogue. Video recording of clinical practice has been proposed to capture observational data and facilitate constructive learning conversations. However, how video influences these conversations remains underexplored. This study investigates how learning conversations are influenced by supervisors' use of video in postgraduate medical education.

This qualitative study explored how supervisors work with video using semi-structured interviews which were conducted with 26 participants between November 2021 and May 2023. Data were analyzed with Reflexive Thematic Analysis.

The sample comprised thirteen supervisors and thirteen trainees from four Anglophone countries, representing diverse specialties. Four themes were generated. (1) Bringing clinical encounters into the conversation: video appeared to enable indirect observation and more accurate focus on specific moments of practice. (2) Inviting trainees to observe themselves: supervisors described encouraging trainee self-reflection and using "objective" evidence to foster consensus. (3) Navigating emotional risks of video review: supervisors noted emotional challenges and described strategies-such as building trust and avoiding assessment framing-to mitigate anxiety. (4) Negotiating control in the learning process: supervisors balanced educational goals with trainee autonomy in context-sensitive ways.

Video can enhance learning conversations, but positive impacts depend on thoughtful implementation. Poor use may increase trainee discomfort or defensiveness, while skillful application can enrich dialogue, shift perspectives, and promote reflection. These findings highlight the complexity of integrating video and the need for considered educational design decisions to optimize its impact on learning conversations.