Menglin Cheng, Yawen Chen, Xingru Yu, Mingyuan Xu, Na Liao, Ying Yang, Xiaofeng Huang
Department of Stomatology, Beijing Friendship Hospital, Capital Medical University, Beijing, China.
Department of Stomatology, Tongzhou District Maternal and Child Health Care Hospital, Beijing, China.
BMC Oral Health. 2025 Aug 19;25(1):1337. doi: 10.1186/s12903-025-06696-7.
This study compared the predictive validity and cost-effectiveness of Cariostat and a reduced Cariogram model for caries risk screening in preschoolers, to guide community-based prevention programs.
A longitudinal study was conducted among 156 children (aged 3-4 years) recruited from two randomly selected kindergartens. Baseline assessments included oral examinations, caregiver questionnaires, and caries-risk stratification using Cariostat and the reduced Cariogram. After 12 months, follow-up examinations recorded caries incidence. Predictive performance was analyzed using sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Cost-effectiveness ratios (CER) and incremental cost-effectiveness ratio (ICER) were calculated per correctly identified moderate-to-high-risk child.
Among 147 children completing follow-up, 50.3% developed new caries (mean dmft increment: 1.41 ± 0.21). The reduced Cariogram showed higher sensitivity (90.24% vs. 73.17%) but similar specificity (43.08% vs. 44.62%) and AUC (0.71 vs. 0.69) compared to Cariostat. The cost per correctly screened moderate-to-high-risk child was 6-fold lower for the reduced Cariogram (¥14.41 vs. ¥82.68), with a negative ICER of -278.3 favoring the reduced Cariogram.
The reduced Cariogram outperformed Cariostat in sensitivity and cost-effectiveness for community caries screening, despite modest specificity. Its lower cost and simpler implementation support use in resource-limited settings.
本研究比较了Cariostat和简化版龋病预测模型在学龄前儿童龋病风险筛查中的预测效度和成本效益，以指导社区预防项目。
从两所随机选取的幼儿园招募了156名3至4岁儿童进行纵向研究。基线评估包括口腔检查、家长问卷，以及使用Cariostat和简化版龋病预测模型进行龋病风险分层。12个月后，随访检查记录龋病发病率。使用灵敏度、特异度和受试者工作特征曲线下面积（AUC）分析预测性能。计算每位正确识别的中高风险儿童的成本效益比（CER）和增量成本效益比（ICER）。
在147名完成随访的儿童中，50.3%出现了新发龋病（平均dmft增量：1.41±0.21）。与Cariostat相比，简化版龋病预测模型显示出更高的灵敏度（90.24%对73.17%），但特异度相似（43.08%对44.62%），AUC也相似（0.71对0.69）。简化版龋病预测模型筛查每位正确识别的中高风险儿童的成本低6倍（14.41元对82.68元），ICER为-278.3，有利于简化版龋病预测模型。
简化版龋病预测模型在社区龋病筛查的灵敏度和成本效益方面优于Cariostat，尽管特异度一般。其较低的成本和更简单的实施方式支持在资源有限的环境中使用。
