Development of an Assessment for Entrustable Professional Activity (EPA) 10: Emergent Patient Management

INTRODUCTION: Medical schools in the United States are encouraged to prepare and certify the entrustment of medical students to perform 13 core entrustable professional activities (EPAs) prior to graduation. Entrustment is defined as the informed belief that the learner is qualified to autonomously perform specific patient-care activities. Core EPA-10 is the entrustment of a graduate to care for the emergent patient. The purpose of this project was to design a realistic performance assessment method for evaluating fourth-year medical students on EPA-10. METHODS: First, we wrote five emergent patient case-scenarios that a medical trainee would likely confront in an acute care setting. Furthermore, we developed high-fidelity simulations to realistically portray these patient case scenarios. Finally, we designed a performance assessment instrument to evaluate the medical student’s performance on executing critical actions related to EPA-10 competencies. Critical actions included the following: triage skills, mustering the medical team, identifying causes of patient decompensation, and initiating care. Up to four students were involved with each case scenario; however, only the team leader was evaluated using the assessment instruments developed for each case. RESULTS: A total of 114 students participated in the EPA-10 assessment during their final year of medical school. Most students demonstrated competence in recognizing unstable vital signs (97%), engaging the team (93%), and making appropriate dispositions (92%). Almost 87% of the students were rated as having reached entrustment to manage the care of an emergent patient (99 of 114). Inter-rater reliability varied by case scenario, ranging from moderate to near-perfect agreement. Three of five case-scenario assessment instruments contained items that were internally consistent at measuring student performance. Additionally, the individual item scores for these case scenarios were highly correlated with the global entrustment decision. CONCLUSION: High-fidelity simulation showed good potential for effective assessment of medical student entrustment of caring for the emergent patient. Preliminary evidence from this pilot project suggests content validity of most cases and associated checklist items. The assessments also demonstrated moderately strong faculty inter-rater reliability.