Clinical Trial Site Weekly Snapshots with Quad Metrics

Site week snapshot data includes core quad metrics: Dropout rate, AE burden, Efficacy lag, and Informative censoring. The dataset targets predicting analysis failure in the next 60 days with label_analysis_fail_next_60d. Author ClarusC64 released it under an MIT license in February 2026.

Use Cases

• Predict label_analysis_fail_next_60d using the core quad metrics of Dropout rate, AE burden, Efficacy lag, and Informative censoring.
• Analyze the relationship between AE burden and Dropout rate across site week snapshots.
• Model the impact of Efficacy lag on the risk of future analysis failure.
• Benchmark predictive models for clinical trial site performance using the provided scorer.py on test data.

Strengths

• Includes four defined core metrics (Dropout rate, AE burden, Efficacy lag, Informative censoring) for analysis.
• Provides a clear target variable, label_analysis_fail_next_60d, for supervised learning tasks.
• Released under a permissive MIT license for broad usage.
• Includes evaluation script (scorer.py) and separate train/test splits for benchmarking.

Limitations

• Unknown row count and column details limit assessment of dataset scale and feature completeness.
• Lack of sample data or file format specifics prevents verification of data structure and quality.
• Reliance on site week snapshots may introduce temporal aggregation biases.

Provenance

Source

Hugging Face, author ClarusC64.

Collection Method

Appears to be aggregated site-level clinical trial metrics, method unknown.

Time Range

null

Freshness

Last updated February 2026.

Geography

null