Bayesian Gamma-Poisson Signal Detection for Drugs & Devices

Upload FAERS or MAUDE data (CSV with product, adverse event columns) to detect disproportionate reporting signals.

Expected columns: product, event

Each row = one adverse event report.

Signal Detection Results

About aers.mobi

Bayesian disproportionality analysis for drug and medical device adverse event reporting using FDA data sources.

Data Sources

  • FAERS — FDA Adverse Event Reporting System (drugs & biologics)
  • MAUDE — Manufacturer and User Facility Device Experience (medical devices)

Statistical Method

  • Prior: 2-component Gamma mixture fitted via EM algorithm
  • Posterior: full Gamma mixture (not EBGM approximation)
  • Signal: EB05 (5th percentile of posterior) > threshold
  • Reference: DuMouchel (1999), American Statistician 53(3)