Data Sheet 1_Development of a nomogram for predicting incident heart failure and all-cause

Data Sheet 1 contains clinical data from a 3-year follow-up study of 440 patients with chronic kidney disease stages 3–5. The dataset, authored by Yuxin Jiang and shared under CC-BY-4.0, was used to develop a nomogram for predicting incident heart failure and all-cause mortality. The study was conducted from November 2020 to September 2024, with a median follow-up of 29 months.

Use Cases

• Training risk prediction models based on identified predictors like age, GDF-15, Lp(a), ferritin, and EQ-5D score.
• Validating clinical nomogram performance using time-dependent AUC metrics reported at 24, 27, 32, and 36 months.
• Conducting decision curve analysis to assess clinical utility across different threshold probabilities.
• Studying the relationship between biochemical markers and cardiovascular outcomes in a CKD cohort.

Strengths

• Includes 440 patient records with a median follow-up of 29 months, providing longitudinal data.
• Model performance is quantitatively reported with a C-index of 0.772 and time-dependent AUC values up to 0.850.
• Key predictors for the nomogram were identified using multiple algorithms (Cox regression, Lasso, Random Forest, XGBoost).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic or institutional bias inherent to recruitment from a single nephropathy center.

Provenance

Source

Yuxin Jiang

Collection Method

Prospective recruitment from a nephropathy center, with collection of clinical, demographic, biochemical, echocardiographic, and quality-of-life data.

Time Range

November 2020 to September 2024, with follow-up over 29 months (IQR 24–35).

Freshness

Last updated 2026-05 08 05:40:38; freshness should be verified.

Geography

null