Minimal Synthetic Data for Type 2 Diabetes Study Reproducibility

A 342.7 KB synthetic dataset generated to reproduce the main analytical results of a clinical study. The data is derived from a retrospective cohort of adult patients with type 2 diabetes mellitus admitted to the Beth Israel Deaconess Medical Center (BIDMC) in Boston, USA, between 2008 and 2019. The synthetic version, created by xiaohan li, preserves the marginal distributions and correlation structures of the original restricted data.

Use Cases

• Testing reproducibility workflows based on the described synthetic data generation method.
• Methodological research on synthetic data quality for clinical studies based on the described correlation structure preservation.
• Educational demonstrations of statistical analysis for type 2 diabetes research based on the described cohort features.

Strengths

• Data is fully de-identified and contains no real patient-level information, addressing privacy concerns.
• Synthetic data preserves the marginal distributions and correlation structures of the original clinical dataset.
• Shared under a permissive CC-BY-4.0 license to support transparency and secondary research.

Limitations

• Row count is unknown, which may limit suitability assessment for specific analytical tasks.
• Column-level documentation is absent; field semantics must be inferred after download.
• The 342.7 KB file size suggests a very limited scope, containing only the minimal data required to reproduce specific results.

Provenance

Source

Synthetically derived from a retrospective cohort at Beth Israel Deaconess Medical Center (BIDMC), Boston, USA.

Collection Method

Generated to preserve marginal distributions and correlation structures of the original, restricted patient data.

Time Range

2008 to 2019

Freshness

Last updated 2026-05-01 09:27:04.

Geography

Boston, Massachusetts, USA