Identifying individuals at risk of needing CKD associated medications in a European kidney disease cohort

Area of Science:

• Nephrology
• Pharmacotherapy
• Clinical Prediction Models

Background:

• Chronic kidney disease (CKD) management involves various pharmacotherapies.
• Accurate prediction of future medication needs can optimize clinical decision-making and patient follow-up schedules.

Purpose of the Study:

• To identify predictors for the future use of specific pharmacotherapies in patients with advanced CKD (stage G4/G5).
• To assess the potential for these predictions to inform follow-up frequency.

Main Methods:

• A prospective cohort study across six European countries analyzed demographic, comorbidity, hospitalization, laboratory, and mortality data.
• Logistic regression models were used to identify variables predicting the use of vitamin D receptor agonists (VDRA), phosphate binders, erythropoiesis-stimulating agents (ESAs), and iron.
• Model performance was evaluated using C-statistics in derivation and validation cohorts.

Main Results:

• The study included 2196 patients with CKD stage G4/G5.
• Combinations of age, diabetes status, iPTH, calcium, hemoglobin, and serum albumin levels predicted medication use (C-statistics ranging from 0.63 to 0.73).
• Sixteen percent of patients had a predicted likelihood of less than 20% for requiring these medications.

Conclusions:

• Patient characteristics can predict the need for erythropoiesis-stimulating agents (ESAs) and phosphate binders within a two-year period in a multi-country CKD cohort.
• This predictive capability holds potential for reducing follow-up frequency in low-risk CKD patients.