An Approach for Personalized Dynamic Assessment of Chronic Kidney Disease Progression Using Joint Model

Area of Science:

• Nephrology
• Biostatistics
• Public Health

Background:

• Chronic kidney disease (CKD) presents significant public health challenges due to its progressive nature.
• Existing methods often fail to capture the individual variability in renal function decline.
• Accurate prediction of CKD progression is crucial for timely intervention and management.

Purpose of the Study:

• To develop a novel, individualized approach for assessing CKD progression.
• To enhance the prediction of progression to hemodialysis treatment.
• To provide a more nuanced understanding of factors influencing renal function in CKD patients.

Main Methods:

• A cohort of 1042 CKD patients (stages 3a-5) was studied over 12 years.
• Clinical measurements were collected and analyzed using joint models.
• Longitudinal and survival analyses were combined to predict hemodialysis progression.

Main Results:

• Estimated glomerular filtration rate (eGFR) is influenced by biochemical markers.
• Serum creatinine, blood pressure, and urinary protein excretion are key predictors of CKD progression.
• The joint model demonstrated predictive versatility across different CKD severities.

Conclusions:

• The novel joint model offers a personalized assessment of renal failure risk.
• This approach improves upon conventional models by integrating longitudinal and survival data.
• Findings support enhanced management and personalized healthcare interventions for CKD patients.