摘要：Objectively structured clinical examinations (OSCEs) are a stressful experience for many health care students and professionals in training. Mock OSCEs have been shown to be beneficial for student OSCE preparation. However, due to their expense and administrative burden students may only get a few opportunities to partake in these. To address this gap in student preparation a series of regularly run totally peer led multi-role practice OSCEs (PrOSCEs) was developed. Fifteen PrOSCEs were run over five-months. A total of 32 second year medical students took part, all of whom were enrolled on the graduate-entry programme at the University of Southampton. In each PrOSCE, 18 participants rotated through the roles of ‘student’, ‘examiner’ and ‘patient’ in six simulated stations designed by their peers. Peer feedback was provided after each station. At the end of the series of PrOSCEs students were asked to fill in an anonymous online feedback survey to assess the usefulness of the PrOSCEs in exam preparation. Twenty-two students responded to the survey. 100% of respondents deemed routine participation either ‘very useful’ or ‘useful’ in preparing for their exam. PrOSCEs were found to improve confidence (mean = 7.9/10, 95% CI 7.4–8.3), expected performance (mean = 7.5/10, 95% CI 6.8–8.2) and help guide revision (mean = 8.3/10, 95% CI 7.6–9.0). Self-perceived teaching performance and confidence in providing feedback was also positively associated with participation. The most beneficial roles were ‘student’ and ‘station creator’. Free-text feedback suggests that the informal setting and regular practice were particularly beneficial. The peer-led nature of the PrOSCEs allows for a low cost, low administrative burden and easy to replicate adjunct or alternative to large scale mock OSCEs. In addition the multi-role aspect of this approach could enhance exam preparation and may also improve aptitude as a clinical teacher. Further studies are required to understand if repeated practice has beneficial implications on OSCE performance.
