Statistical Power for Detecting HbA1c Fraud in Clinical Trials

Philippe P. Hujoel authored a dataset analyzing the sensitivity of a statistical method for detecting simulated fraud in clinical trial data. The dataset, last updated on April 15, 2026, is a 5.5 KB Excel file. It models scenarios where 50% to 90% of the worst HbA1c scores in an intervention arm are replaced by a best-responder value.

Use Cases

• Calculate required sample sizes for fraud detection based on simulated HbA1c score manipulation.
• Evaluate the power of the 3-sigma lnCVR statistical test for different fraud severity levels.
• Benchmark fraud detection sensitivity across various trial arm sizes.
• Model the impact of data fabrication on clinical trial outcome reliability.

Strengths

• Dataset is openly licensed under CC-BY-4.0.
• File size is 5.5 KB, indicating a focused and manageable analysis.
• The description specifies a clear simulation scenario (50-90% worst HbA1c scores replaced).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is a simulation; its applicability to real-world fraud detection requires validation.

Provenance

Source

figshare

Freshness

Last updated 2026-04-15 17:45:10; freshness should be verified.