Authors
Junjun Hou, Xiuyu He, Fei Gao, Ting Ren, Lijin Ren, Xianguo Wang
Published in
Risk management and healthcare policy. Volume 19. Pages 607198. Epub Jul 16, 2026.
Abstract
To develop and validate a user-friendly nomogram for predicting the risk of esophageal squamous cell carcinoma (ESCC) and high-grade intraepithelial neoplasia (HGIN), designed for initial screening settings while addressing variable complexity and class imbalance in traditional models.
Based on a screening cohort of 23,257 participants from Tai'an, Shandong, between 2024 and 2025 (positive rate: 1.54%), a 1:10 case-control sampling method was applied to address the low event rate (positive rate: 1.54%) and correct class imbalance. Predictors were initially screened using LASSO regression with 10-fold cross-validation (λ.min criterion) and further refined via multivariable logistic regression to establish the final model, which was presented as a nomogram and evaluated via internal split-sample validation.
Seven easily accessible predictors were identified: age, sex, education level, BMI, smoking history, hot-food consumption, and family history of esophageal cancer. The model showed strong discriminatory performance, with an AUC of 0.823 (95% CI: 0.798-0.848) in the training set and 0.835 (95% CI: 0.805-0.865) in the internal validation set. Calibration curves indicated high consistency between predicted and observed risks. Decision curve analysis demonstrated net clinical benefit across risk thresholds of 0-0.6.
The proposed simplified nomogram demonstrates promising potential for risk stratification in primary ESCC/HGIN screening. However, prospective external validation in diverse cohorts is necessary before its large-scale clinical implementation.
PMID:
42488795
Bibliographic data and abstract were imported from PubMed on 23 Jul 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 6
- Comments 0