Development of a digital peer-feedback tool for clinical skills training: a pilot study with second-year medical students

Providing structured feedback in medical education is essential for skill development. However, limitations such as increasing student numbers, reducing trainer availability, and a lack of systematic feedback hinder effective learning. Digital peer-feedback tools offer a potential solution by enabling interactive and scalable assessment. This study introduces Medifeeding, a video-based peer-feedback digital tool designed to enhance clinical skills training among medical students. This study employed an exploratory convergent mixed-methods pilot feasibility design. Medifeeding was developed following Moonen's 3-Space model and tested with 31 s-year medical students at & Ccedil;anakkale Onsekiz Mart University. Data sources: a post-use online form (educational benefit yes/no, a forced ranking of 10 attributes, and a 5-point overall rating), semistructured interviews (n = 24), and one focus group (n = 5). Quantitative analyses included frequency analysis, the Friedman test with Kendall's W and Wilcoxon signed-rank tests (Holm-adjusted). Qualitative data were analyzed thematically (Strauss's framework). Strands were integrated in a joint display to generate meta-inferences on acceptability and usability. 90% (28/31) reported educational benefit. The overall rating averaged 4.13/5 (90% rated 4-5). For the feature ranking, the Friedman test showed an overall difference (**chi 2(9) = 86.162, p < 0.001; Kendall's W = 0.309). The highest-priority attribute was 'educationally beneficial' (mean rank approximate to 2.13; median 1), followed by 'offering a different experience' (approximate to 4.35) and 'digitally user-friendly' (approximate to 4.84); post-hoc Wilcoxon (Holm) confirmed multiple pairwise differences. Qualitatively, six themes emerged: (1) educational benefits, (2) peer interaction, (3) moderation and trust, (4) new usage scenarios, (5) awareness and motivation, and (6) technical infrastructure. Students highlighted self-assessment and peer learning; concerns included misinformation and technical limitations. Gamification and AI-assisted moderation were suggested as potential improvements. This study underscores the exciting possibilities of Medifeeding as a digital peer-feedback system that can greatly aid the clinical skill mastery of medical students. In this pilot with second-year students, participants reported educational benefit and positive usability while also noting concerns about the validity of peer reviews, the risk of misinformation, and variable engagement. Overall, our findings indicate that Medifeeding is a potential tool to support structured peer feedback in clinical skills training; however, larger and longer-term studies are needed to evaluate its impact on performance and learning outcomes, usability, and feedback validation.