耳鼻咽喉科住院医师手术操作表现的客观评估

文献信息

中文摘要

目的： 外科医生的技术能力是手术结局的关键预测因素，然而住院医师的表现常通过直接观察进行评估，缺乏结构化、预先设定的标准，这可能存在偏倚且不可靠。本研究旨在在耳鼻咽喉科住院医师培训中针对多种手术实施结构化评估工具，评估其可行性，并评价其对住院医师表现的影响。
方法： 每个住院医师年级选择两项手术。从文献中采用七种经过验证的工具，并新开发第八种工具；每种工具均包含任务特异性核查表（TSC）和整体评定量表（GRS）。住院医师接受两轮评估，每轮由三名医师评估并加一次自我评价。主要结局是八种评估表的实施情况以及在常规实践中多评估者、多手术评估的可行性。次要结局是对表现的影响，以两轮之间TSC和GRS评分的变化以及教师-自我评价一致性来衡量。
结果： 全部八种评估表均已实施并数字化；八项手术中有六项评估可行，因为内镜鼻窦手术和儿童支气管镜检查符合条件病例过少。各单项手术在两轮之间未发现显著差异，但观察到无显著性的改善趋势。汇总的自我评价随时间显著改善。自我评价与教师评价无显著差异，但统计效能有限，无法得出确切结论。低年资住院医师改善最明显，而高年资住院医师表现出天花板效应。
结论： 本初步研究证明在耳鼻咽喉科住院医师培训中开展结构化、多评估者手术评估的可行性。使用多名评估者可限制评分者偏倚，并将使评估者间信度检验成为可能，而该检验在当前阶段尚不可行。这些初步结果支持开展更大规模的多中心研究，在更广泛实施前纳入信度和效度检验。

英文摘要

PURPOSE: A surgeon's technical competence is a key predictor of surgical outcomes, yet residents' performance is often assessed by direct observation without structured, predefined criteria, which can be biased and unreliable. This study aimed to implement structured assessment tools across multiple procedures in an ENT residency, assess their feasibility, and evaluate their impact on resident performance.
METHODS: Two procedures per residency year were selected. Seven validated tools were adopted from the literature and an eighth was newly developed; each comprised a Task-Specific Checklist (TSC) and a Global Rating Scale (GRS). Residents were assessed over two rounds, each by three physicians plus a self-evaluation. The primary outcomes were implementation of the eight grids and the feasibility of multi-evaluator, multi-procedure assessment in routine practice. The secondary outcome was impact on performance measured as the change in TSC and GRS scores between rounds and the faculty-self agreement.
RESULTS: All eight grids were implemented and digitized; assessment was feasible for six of eight procedures, as endoscopic sinus surgery and pediatric bronchoscopy yielded too few eligible cases. No significant between-round differences were found for individual procedures, though non-significant improvement trends were observed. Pooled self-assessments improved significantly over time. Self- and faculty assessments did not differ significantly, though limited statistical power precludes firm conclusions. Junior residents improved most, whereas senior residents showed a ceiling effect.
CONCLUSION: This pilot study demonstrates the feasibility of structured, multi-evaluator surgical assessment in an ENT residency. Using multiple evaluators limits rater bias and will enable inter-rater reliability testing, not feasible at this stage. These preliminary results support a larger, multi-center study incorporating reliability and validity testing before broader implementation.