Development and validation of a chronic kidney disease progression model using patient-level simulations

Area of Science:

• Nephrology
• Biostatistics
• Epidemiology

Background:

• Existing chronic kidney disease (CKD) models lack comprehensiveness, focusing narrowly on kidney failure and cardiovascular (CV) events.
• There is a need for advanced models to predict a wider spectrum of CKD complications and patient outcomes.

Purpose of the Study:

• To develop and validate a comprehensive CKD progression model (CKD-PM).
• The CKD-PM aims to simulate risks for CKD progression, major complications, and mortality.

Main Methods:

• Literature reviews identified risk factors and existing algorithms for CKD outcomes.
• A patient-level, state-transition model was built using data from CKD registries.
• Model validation included internal and external comparisons of predicted versus observed outcomes.

Main Results:

• The CKD-PM showed satisfactory and robust modeling properties upon validation.
• Accurate prediction of all-cause and CV mortality was achieved.
• The model accurately predicted estimated glomerular filtration rate (eGFR) changes and kidney replacement therapy (KRT) events.

Conclusions:

• The developed CKD-PM is a reliable tool for predicting diverse CKD outcomes.
• This comprehensive model can improve risk stratification and management strategies for patients with CKD.