Document resource
Objectives To evaluate whether integration of AI-generated post-consultation feedback in social robotic VP interactions improves medical students’ clinical performance.Methods We conducted a quasi-experimental study with 115 sixth-semester medical students at Karolinska Institutet (KI), Sweden. During their clinical rotation within rheumatology, students were allocated to either receive (n=61) or not receive (n=54) AI-generated feedback following interactions with a Social AI-enhanced Robotic Interface (SARI). Students in both arms participated in case-specific follow-up seminars led by clinical experts following each VP encounter. Clinical performance was assessed during objective structured clinical examination (OSCE) with a standardised patient, evaluated by consultant rheumatologists using a 10-point rubric.Results Students receiving AI-generated feedback achieved significantly higher total OSCE scores (7.39 ± 0.86 versus 6.68 ± 1.04 points; mean difference: 0.70; 95% CI: 0.35–1.06; p<0.001; Cohen’s d=0.74). Domain-specific analysis revealed significant improvement only in generic medical history, also after Bonferroni correction (2.46 ± 0.65 versus 2.03 ± 0.79 points; p=0.004). Pass rates were significantly higher in the feedback group (96.7% versus 79.6%; OR: 7.55; 95% CI: 1.51–72.2; p=0.006), with a number needed to treat of six.Abstract PO:12:310 Figure 1Conclusions AI-generated feedback following social robotic VP interactions significantly improved medical students’ clinical performance in standardised examination. These findings support integrating AI feedback systems in VP platforms for clinical skill training in rheumatology, while highlighting the importance of targeted, competency-specific feedback design.