Development and validation of a nomogram for predicting advanced chronic kidney disease

The objective of this study was to develop and validate a nomogram for predicting advanced chronic kidney disease (CKD) through the utilization of routine clinical parameters. To achieve this, we integrated internal and external datasets, employed LASSO and logistic regression to identify independent predictors of advanced CKD for nomogram construction, conducted Bootstrap validation on the training set, and evaluated discrimination, calibration, and net clinical benefit utilizing both internal and external test sets. The results identified six predictors: serum creatinine, age, hematocrit, hypertension, serum ALT, and serum LDH. Furthermore, it was demonstrated that the nomogram exhibited excellent discrimination (AUC: 0.961/0.965/0.939) and good calibration (Brier score: 0.058/0.056/0.073) across the training, internal, and external sets, alongside significant net clinical benefit within the 0-1 threshold. Consequently, it was concluded that this robust and clinically practical nomogram facilitates accurate, individualized predictions of advanced CKD risk, thereby supporting precision kidney disease management.