實證醫學之客觀結構式臨床測驗評量信效度評估初探

The Validity and Reliability of using OSCE as an Instrument to Assess EBM Competency-A Preliminary Analysis

黃靜宜(Chin-Yi Huang) ； 廖以誠(Yi-Cheng Liao) ； 曹龍彥(Lon-Yen Tsao)

《The Changhua Journal of Medicine》 11卷2期 (2013/12) Pp. 103-111

摘要

醫學教育機構推動終身學習的教育及多元評量，促進醫學教育品質提昇。所以，畢業後一般醫學訓練將實證醫學列為學員終身學習模式之核心能力之一。然而實證醫學的學習評量仍多以學習者自我評量的方式居多。教育方法及學習評量的同時改善，才能達到品質提升目的。本研究目的，發展實證醫學客觀結構式臨床測驗評量工具，做為國內實證醫學客觀結構式臨床測驗評量工具發展之參考。畢業後一般醫學訓練學員12位參與實證醫學客觀結構式臨床測驗，每位學員均接受前測及後測各1次。實證醫學資深教師群發展實證醫學-客觀結構式臨床測驗案例。學員有16分鐘去進行實證醫學-OSCE的測驗。計算組內相關係數分析檢驗考官間評分的一致性，檢核表測驗品質則以信度分析、相關分析及決斷值分析等項目分析來檢測。2位考官評分者間的一致性組內相關係數值為0.776。檢核表內15個題項的整體量表信度值為0.894。有3個題目與整體量表總分之積差相關係數及題項高、低分組的差異性檢驗上，未達顯著水準，代表這三題的鑑別力及內部一致性不佳。本研究首次嘗試設計16分鐘的實證醫學-客觀結構式臨床測驗評估學員實證醫學四大能力，藉由結構完整、信效度良好的評估工具，促進考生的學習進步方向。

關鍵字

實證醫學 ； 客觀結構式臨床測驗

並列摘要

The Accreditation Council for Graduate Medical Education highlights the concepts of lifelong learning and multiple assessments in order to elevate the quality of medical education. For that reason, evidence-based medicine is listed as a core competency of lifelong learning in postgraduate training programs. However, most assessments of evidence-based medicine are self-assessed by the learners. The quality of medical education should be raised by way of improvement not only in educational methodologies but also in assessment methods. The aim of this study was to develop an assessment instrument using the Objective Structured Clinical Examination to assess evidence-based medicine competencies. Twelve residents on a postgraduate training program participated in the Objective Structured Clinical Examination. Each trainee attended a pre-exam and a post-exam. The examination scenario was developed by senior teachers in evidence-based medicine. The residents were given 16 minutes to complete the Evidence-based Medicine-Objective Structured Clinical Examination. The intraclass correlation coefficient was used to examine the inter-rater agreement. The assessment quality of the checklist was examined by reliability analysis, correlation analysis and critical ratio analysis. The intraclass correlation coefficient, which indicates the agreement of the two raters, was 0.776. The internal reliability of the global measure of the 15-item checklist was 0.894. Three out of fifteen of the productmoment correlation coefficients of individual checklist items with the global rating score did not reach a significant level, which indicates that the discrimination and internal consistency for these three items were not good enough. To our knowledge, this is the first study to attempt to design a 16-minute Evidence-based Medicine-Objective Structured Clinical Examination. This comprehensive structured assessment instrument, with good reliability and validity, was proved to be capable of assessing students' four competencies in evidence-based medicine as well as facilitating the learning of the students.

並列關鍵字

Evidence-based Medicine ； Objective Structured Clinical Examination