Data Sheet 1_Longitudinal trajectories of urinary albumin-to-creatinine ratio and risk of

A retrospective cohort study of 3,101 Chinese patients with type 2 diabetes and baseline UACR <30 mg/g, with regular follow-up from March 2018 to October 2024. The dataset, authored by Qian Chen and shared under a CC-BY-4.0 license, includes clinical data used to identify three distinct UACR trajectory patterns and their association with proteinuria risk using multivariable Cox regression.

Use Cases

• Modeling urinary albumin-to-creatinine ratio (UACR) trajectories over time using group-based trajectory modeling (GBTM).
• Identifying key clinical features associated with UACR progression using feature selection algorithms like Boruta.
• Predicting proteinuria risk categories based on patient trajectories using multi-class classification models like LightGBM.
• Interpreting feature contributions to model predictions for distinct UACR trajectories using SHAP values.

Strengths

• Includes 3,101 patient records from a single-center cohort.
• Covers a longitudinal time span from March 2018 to October 2024.
• Analysis identifies 11 significant clinical features, including baseline UACR, Hb, BMI, and waist circumference.
• Provides hazard ratios for proteinuria risk across identified trajectory groups.

Limitations

• The dataset is small at 146.0 KB, suggesting limited raw data or a summary document.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data is from a single center in Ningbo, China, which may limit generalizability.

Provenance

Source

Electronic medical records from the Ningbo Metabolic Management Center (MMC) subcenter.

Collection Method

Retrospective cohort study.

Time Range

March 2018 to October 2024

Freshness

Last updated 2026-06-02 04:21:11; freshness should be verified.

Geography

Ningbo, China