Face yourself! - learning progress and shame in different approaches of video feedback: a comparative study

Feedback is a crucial part of medical education and with on-going digitalisation, video feedback has been increasingly in use. Potentially shameful physician-patient-interactions might particularly benefit from it, allowing a meta-perspective view of ones own performance from a distance. We thus wanted to explore different approaches on how to deliver specifically video feedback by investigating the following hypotheses: 1. Is the physical presence of a person delivering the feedback more desired, and associated with improved learning outcomes compared to using a checklist? 2. Are different approaches of video feedback associated with different levels of shame in students with a simple checklist likely to be perceived as least and receiving feedback in front of a group of fellow students being perceived as most embarrassing?

Second-year medical students had to manage a consultation with a simulated patient. Students received structured video feedback according to one randomly assigned approach: checklist (CL), group (G), student tutor (ST), or teacher (T). Shame (ESS, TOSCA, subjective rating) and effectiveness (subjective ratings, remembered feedback points) were measured. T-tests for dependent samples and ANOVAs were used for statistical analysis.

n = 64 students could be included. Video feedback was in hindsight rated significantly less shameful than before. Subjectively, there was no significant difference between the four approaches regarding effectiveness or the potential to arise shame. Objective learning success showed CL to be significantly less effective than the other approaches; additionally, T showed a trend towards being more effective than G or ST.

There was no superior approach as such. But CL could be shown to be less effective than G, ST and T. Feelings of shame were higher before watching one's video feedback than in hindsight. There was no significant difference regarding the different approaches. It does not seem to make any differences as to who is delivering the video feedback as long as it is a real person. This opens possibilities to adapt curricula to local standards, preferences, and resource limitations. Further studies should investigate, whether the present results can be reproduced when also assessing external evaluation and long-term effects.