A Risk Prediction Model for New-Onset Chronic Kidney Disease in the Elderly

Worsening renal function poses a significant health risk to elderly individuals. This study aimed to construct a simple risk prediction model for new-onset chronic kidney disease (CKD) among elderly populations.

In this retrospective cohort study, 5,416 elderly residents (aged ≥65 years) who underwent physical examinations as part of the National Basic Public Health Service project at least twice between January 2017 and July 2021 were included. The endpoint was new-onset CKD, defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 during the follow-up period. Predictors of new-onset CKD were selected using multivariable Cox regression and a stepwise approach. A risk prediction model based on the selected predictors was constructed and evaluated using the concordance index (C-index) and area under curve (AUC). External validation was conducted to verify the model's performance.

During the median follow-up period of 2.3 years, the incident of new-onset CKD was 20.1% (n = 1,088). Age, female gender, diabetes, elevated triglyceride levels, and baseline eGFR were selected as predictors. The model demonstrated good predictive performance across the cohort, with a C-index of 0.802. The AUCs for 2-year, 3-year, and 4-year predictions were 0.831, 0.829, and 0.839, respectively. External validation confirmed the model's efficacy, with a 2-year AUC of 0.735.

This study developed a simple yet effective risk prediction model for new-onset CKD among elderly populations. The model facilitates prompt identification of elderly individuals at risk of renal function decline in primary care, enabling timely interventions.