文献速递牙体牙髓病儿童口腔

牙科虚拟标准化病人在牙髓病临床推理与决策中的初步评估:表面效度与结构效度

Preliminary Evaluation of a Dental Virtual Standardized Patient in Endodontic Clinical Reasoning and Decision-Making: Face and Construct Validity.
PubMed 文献 · Int Endod J · 2026-07-28

本研究开发并评估了一种整合自然语言意图识别、高保真触觉模拟器和椅旁视觉的牙科虚拟标准化病人(D-VSP),用于牙髓病临床推理与决策培训。重庆医科大学口腔医学院149名学生按自报临床经验分为四组,完成无并发症冠折(UCF)和症状性不可逆牙髓炎(SIP)两个虚拟病例并填写可用性与保真度问卷。结果显示各组对D-VSP的可用性和保真度评价均高,SIP测试总分及临床检查、诊断部分存在显著组间差异,UCF仅临床检查部分有显著差异,支持D-VSP的表面效度和结构效度。

要点

  • D-VSP整合自然语言识别与触觉模拟,可用性和保真度获学生高评价。
  • SIP病例总分及临床检查、诊断部分组间差异显著,提示结构效度。
  • UCF病例仅临床检查部分组间差异显著,区分度有限。

核心结论

  • D-VSP在牙髓病临床推理教学中具有表面效度和结构效度,可作为教学与评估工具。

研究设计与样本

  • 开发D-VSP并纳入重庆医科大学口腔医学院学生,按自报临床经验分为四组。
  • 共149名学生完成UCF和SIP两个虚拟标准化病人病例及两份5点Likert问卷。
  • 统计分析采用单因素方差分析或Kruskal-Wallis检验,Bonferroni校正,显著性水平p<0.05,效应量以eta平方(η2)报告。

主要发现

  • 所有组对D-VSP的可用性和保真度评分均高。
  • SIP测试总分组间差异显著(H=10.914,p=0.012,η2=0.057)。
  • SIP测试临床检查部分组间差异显著(H=8.219,p=0.042,η2=0.037)。
  • SIP测试诊断部分组间差异显著(H=22.063,p<0.001,η2=0.136)。
  • UCF病例仅临床检查部分组间差异显著(H=8.496,p=0.037,η2=0.039)。

临床怎么用

  • D-VSP可用于牙髓病临床推理与决策的培训和形成性/终结性评估。
  • SIP病例在诊断部分区分度较高(η2=0.136),适合用于评估诊断能力。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

AIM: To develop and evaluate a dental virtual standardized patient (D-VSP) integrating natural-language intent recognition, a high-fidelity haptic simulator, and chairside visuals, to establish its face validity as a training tool for endodontic clinical reasoning and decision-making, and to examine its construct validity as a summative assessment instrument. METHODOLOGY: The D-VSP delivered a structured clinical workflow comprising six sections: history-taking, clinical examination, documentation, ancillary testing, diagnosis, and treatment planning. Dental students from the Stomatology College of Chongqing Medical University were classified into four pre-specified groups according to their self-reported clinical experience. All participants completed two virtual standardized patient cases: uncomplicated crown fracture (UCF) and symptomatic irreversible pulpitis (SIP). They then filled out two questionnaires assessing perceived usability and fidelity on 5-point Likert scales. Statistical analysis employed one-way analysis of variance or the Kruskal-Wallis test, with Bonferroni correction applied for multiple comparisons. The significance level was set at p < 0.05, and effect sizes were reported as eta-squared (η2). RESULTS: A total of 149 students were recruited (data recovery rate: 96.64%). All groups rated the D-VSP highly for its usability and fidelity. SIP test scores differed significantly among the groups (H = 10.914, p = 0.012, η2 = 0.057). In addition, significant differences were found in the clinical examination section (H = 8.219, p = 0.042, η2 = 0.037) and the diagnosis section (H = 22.063, p < 0.001, η2 = 0.136) of the SIP test. For the UCF case, only the clinical examination section showed a significant difference (H = 8.496, p = 0.037, η2 = 0.039). CONCLUSION: This study established face and construct validity for the D-VSP, supporting its potential as a tool for teaching and assessing clinical reasoning in endodontics.

来源:Int Endod J,查看原文。版权归原作者所有。

引用

  1. Yuan Z, Zhu Q, Ren Y, et al. Preliminary Evaluation of a Dental Virtual Standardized Patient in Endodontic Clinical Reasoning and Decision-Making: Face and Construct Validity. Int Endod J. 2026. PMID: 42522165. doi:10.1111/iej.70223

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。