牙周病自报筛查工具诊断准确性的外部验证试验
本研究在香港普通人群中随机抽取384名受试者,采用改良CDC/AAP问卷结合人口学与生活方式因素构建筛查模型,以2017世界研讨会牙周病分类为参照进行外部验证。模型检测牙周炎的AUROC为0.888,敏感度91.1%、特异度90.0%;检测III/IV期牙周炎的AUROC为0.899,敏感度88.8%、特异度77.2%。结论认为该模型在独立人群中具有可接受的准确性。
要点
- 联合CDC/AAP问卷与人口学及生活方式因素的模型对牙周炎筛查AUROC达0.888。
- 该模型检测III/IV期牙周炎的AUROC为0.899,敏感度88.8%、特异度77.2%。
- 外部验证支持该筛查工具在不同社区场景中推广应用的潜力。
核心结论
- 外部验证的筛查模型在独立人群中识别牙周炎患者具有可接受的准确性,支持其在不同社区场景中的潜在适用性。
研究设计与样本
- 研究采用自填问卷与全口牙周及影像学检查,在香港普通人群中系统随机抽取代表性队列进行外部验证。
- 共384名受试者参与验证,包括牙周健康者6.2%、牙龈炎19.3%、I/II期牙周炎19.3%和III/IV期牙周炎55.2%。
- 使用改良CDC/AAP问卷,以2017世界研讨会牙周病分类的病例定义作为参照标准。
- 采用多变量逻辑回归分析准确性,并计算受试者工作特征曲线下面积(AUROC)。
主要发现
- 整合CDC/AAP问卷与人口学及生活方式因素的筛查模型在检测牙周炎和重度牙周炎方面表现最佳。
- 检测牙周炎时,模型AUROC为0.888,敏感度91.1%,特异度90.0%。
- 检测III/IV期牙周炎时,模型AUROC为0.899,敏感度88.8%,特异度77.2%。
关键数值与推荐
- 牙周炎检测:AUROC 0.888,敏感度91.1%,特异度90.0%。
- III/IV期牙周炎检测:AUROC 0.899,敏感度88.8%,特异度77.2%。
临床怎么用
- 该问卷联合模型可用于社区人群牙周炎初步筛查,辅助识别需进一步牙周检查的个体。
- 模型对重度牙周炎具有较高敏感度,有助于早期发现高风险患者并促进早期牙周 care。
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 externally validate a previously developed time- and cost-effective screening tool for screening subjects with different periodontal conditions by combining various questionnaires, demographic data and lifestyle factors, providing a reliable basis for future efforts to promote public awareness of oral/periodontal health and early periodontal care. METHODS: The study consisted of a self-administered questionnaire and a full-mouth periodontal and radiographic examination. External validation was conducted using a representative cohort systematically and randomly selected from the general population of Hong Kong. A modified questionnaire originally proposed by the Centers for Disease Control and Prevention/American Academy of Periodontology (CDC/AAP) was given to participants, and the results were tested against the periodontal case definition based on the 2017 World Workshop Classification of Periodontal Diseases. Accuracy was analysed by multivariate logistic regression, and the area under the receiver operating characteristic curve (AUROC) was calculated. RESULTS: A total of 384 subjects participated in this validation study, including individuals with periodontal health (6.2%), gingivitis (19.3%), Stage I/II periodontitis (19.3%) and Stage III/IV periodontitis (55.2%). The screening model, which integrated the CDC/AAP questionnaire with demographic and lifestyle factors, showed the best performance in detecting both periodontitis and severe periodontitis. For periodontitis detection, the model achieved an AUROC of 0.888, with a sensitivity of 91.1% and specificity of 90.0%. For Stage III/IV periodontitis, it showed an AUROC of 0.899, with a sensitivity of 88.8% and specificity of 77.2%. CONCLUSION: This externally validated model demonstrates acceptable accuracy in identifying periodontitis patients in independent populations, supporting its potential applicability across diverse community settings.
引用
- Deng K, Lam SLT, Kung JCK, et al. The External Validation Trial on the Diagnostic Accuracy of Self-Reported Measures of Periodontal Disease. J Clin Periodontol. 2026. PMID: 42834595. doi:10.1111/jcpe.70208
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。