Shikino Kiyoshi, Nishizaki Yuji, Fukui Sho, Kataoka Koshi, Yokokawa Daiki, Shimizu Taro, Yamamoto Yu, Nagasaki Kazuya, Kobayashi Hiroyuki, Tokuda Yasuharu
Department of Community-Oriented Medical Education, Graduate School of Medicine, Chiba University, 1-8-1, Inohana, Chu-ou-ku, Chiba, 2608670, Japan, 81 43-222-7171.
Department of General Medicine, Chiba University Hospital, Chiba, Japan.
JMIR Med Educ. 2025 Jul 31;11:e72640. doi: 10.2196/72640.
Traditional assessments of clinical competence using multiple-choice questions (MCQs) have limitations in the evaluation of real-world diagnostic abilities. As such, recognizing non-verbal cues, like tachypnea, is crucial for accurate diagnosis and effective patient care.
This study aimed to evaluate how detecting such cues impacts the clinical competence of resident physicians by using a clinical simulation video integrated into the General Medicine In-Training Examination (GM-ITE).
This multicenter cross-sectional study enrolled first- and second-year resident physicians who participated in the GM-ITE 2022. Participants watched a 5-minute clinical simulation video depicting a patient with acute pulmonary thromboembolism, and subsequently answered diagnostic questions. Propensity score matching was applied to create balanced groups of resident physicians who detected tachypnea (ie, the detection group) and those who did not (ie, the non-detection group). After matching, we compared the GM-ITE scores and the proportion of correct clinical simulation video answers between the two groups. Subgroup analyses assessed the consistency between results.
In total, 5105 resident physicians were included, from which 959 pairs were identified after the clinical simulation video. Covariates were well balanced between the detection and non-detection groups (standardized mean difference <0.1 for all variables). Post-matching, the detection group achieved significantly higher GM-ITE scores (mean [SD], 47.6 [8.4]) than the non-detection group (mean [SD], 45.7 [8.1]; mean difference, 1.9; 95% CI, 1.1-2.6; P=.041). The proportion of correct clinical simulation video answers was also significantly higher in the detection group (39.2% vs 3.0%; mean difference, 36.2%; 95% CI, 32.8-39.4). Subgroup analyses confirmed consistent results across sex, postgraduate years, and age groups.
Overall, this study revealed that detecting non-verbal cues like tachypnea significantly affects clinical competence, as evidenced by higher GM-ITE scores among resident physicians. Integrating video-based simulations into traditional MCQ examinations enhances the assessment of diagnostic skills by providing a more comprehensive evaluation of clinical abilities. Thus, recognizing non-verbal cues is crucial for clinical competence. Video-based simulations offer a valuable addition to traditional knowledge assessments by improving the diagnostic skills and preparedness of clinicians.
使用多项选择题（MCQ）对临床能力进行的传统评估在评估实际诊断能力方面存在局限性。因此，识别诸如呼吸急促等非语言线索对于准确诊断和有效的患者护理至关重要。
本研究旨在通过使用纳入普通内科住院医师培训考试（GM-ITE）的临床模拟视频，评估检测此类线索如何影响住院医师的临床能力。
这项多中心横断面研究纳入了参加2022年GM-ITE的一年级和二年级住院医师。参与者观看了一段5分钟的临床模拟视频，视频中描绘了一名急性肺血栓栓塞患者，随后回答诊断问题。应用倾向评分匹配法创建了检测到呼吸急促的住院医师平衡组（即检测组）和未检测到呼吸急促的住院医师平衡组（即非检测组）。匹配后，我们比较了两组的GM-ITE分数以及临床模拟视频答案的正确比例。亚组分析评估了结果之间的一致性。
总共纳入了5105名住院医师，临床模拟视频后识别出959对。检测组和非检测组之间的协变量平衡良好（所有变量的标准化平均差<0.1）。匹配后，检测组的GM-ITE分数（均值[标准差]，47.6[8.4]）显著高于非检测组（均值[标准差]，45.7[8.1]；平均差，1.9；95%可信区间，1.1 - 2.6；P = 0.041）。检测组临床模拟视频答案的正确比例也显著更高（39.2%对3.0%；平均差，36.2%；95%可信区间，32.8 - 39.4）。亚组分析证实了不同性别、研究生年级和年龄组的结果一致。
总体而言，本研究表明，检测到诸如呼吸急促等非语言线索会显著影响临床能力，住院医师中较高的GM-ITE分数证明了这一点。将基于视频的模拟纳入传统的MCQ考试，通过对临床能力进行更全面的评估，增强了对诊断技能的评估。因此，识别非语言线索对临床能力至关重要。基于视频的模拟通过提高临床医生的诊断技能和准备程度，为传统知识评估提供了有价值的补充。
