부산대학교 도서관

Introduction: Simulation training is becoming increasingly prevalent in medical education. Dissemination and implementation of simulation is limited by financial cost or validity of the simulation. Few low-cost bronchoscopy models are readily available. We sought to create and provide validity evidence for a low-fidelity model to teach bronchoscopy.Methods: We developed a low-fidelity bronchoscopy model from low-cost, easily procured materials. In a nonrandomized study, we compared three groups defined by number of career bronchoscopies performed: Novice (0–5), Intermediate (6–49), and Expert (≥ 50). Each participant underwent written examination as well as performance evaluation on the simulator. Tasks were graded by an independent evaluator with regard to time to completion, correct identification, scope positioning, and scope airway trauma. In accordance with Messick’s standard framework, evidence of validity required that each group performs significantly different than the others with Experts performing best and Novices worst; Experts required a score > 65%. Following the physical simulation, Experts were surveyed using a 5-point Likert scale to evaluate the ability of the simulator to teach as well as “realism.”Results: 37 physicians completed the simulation: 14 Novice, 14 Intermediate, and 9 Expert. Expert, Intermediate, and Novice physical simulation scores were significantly different between groups and were ordinally arranged with Experts performing best, with Expert scores on average exceeding 65% (pExpert-Intermediate < 0.01, pExpert-Novice < 0.01, pIntermediate-Novice < 0.01). Expert written exam scores were significantly better than Novice or Intermediate (pExpert-Intermediate < 0.05, pExpert-Novice < 0.01, pIntermediate-Novice = 0.42), but Experts did not perform at the satisfactory cutoff of 65% required to validate the written exam alone. When written exams were combined with physical simulation scores, validation criteria were again met satisfactorily. On 14 of 16 bronchoscopic skills, study participants generally reported increases in self-assessed capability after completion of the simulation. One task, identification of segmental anatomy, was generally identified as a weak area across the whole cohort. With regard to content, the mean scores given by Experts for all six parameters of model effectiveness were in the range of 4–5/5, indicating that Experts on average felt moderate to strongsatisfaction with the model’s effectiveness as a teaching tool. Three of seven descriptive features were rated in the neutral to moderately satisfying with regard to realism, while the remaining four features were moderately to strongly satisfying with regard to realism.Conclusions: We have designed a low-cost bronchoscopy simulator that has demonstrated the ability to distinguish between experts and non-experts; it has been found by experts to have appropriate content. Subjects had individualized experiences with regard to perceived changes in self-assessed capability. This will impact bronchoscopy simulation, medical education, as well as education costs.