faers

Modern biologics, such as immune checkpoint inhibitors, exhibit complex toxicity profiles that are often underrepresented in pre-market clinical trials. While the FAERS database serves as a critical resource for real-world safety surveillance, its intricate relational structure and data inconsistencies pose significant barriers to large-scale epidemiological analyses.

To address these challenges, we developed faers, an end-to-end, reproducible framework for precision pharmacovigilance. The package streamlines the entire workflow—from raw data acquisition and rigorous preprocessing to signal detection—empowering researchers to transform vast spontaneous reporting data into actionable clinical insights.

Key Features

• 📥 Data Acquisition: Automated downloading and parsing of FAERS quarterly data (supporting both ASCII and XML formats).

• 🛠️ Rigorous Preprocessing:Multi-quarter data merging and deduplication logic for data quality control.

• 🔍 Terminology Standardization: Integration with MedDRA and RxNorm for precise mapping of drugs and adverse events.

• 📊 Advanced Signal Detection: Comprehensive disproportionality analysis, including ROR, PRR, BCPNN, and EBGM.

• ⚡ High-Performance Computing: BiocParallel integration for memory-efficient, parallelized processing.

• 🌐 Knowledge Integration: Support for Athena drug vocabularies and Standardized MedDRA Queries (SMQ) for mechanism-driven research.

Installation

To install from Bioconductor, use the following code:

if (!requireNamespace("BiocManager", quietly = TRUE)) {
    install.packages("BiocManager")
}
BiocManager::install("faers")

You can install the development version of faers from GitHub with:

if (!requireNamespace("pak", quietly = TRUE)) {
    install.packages("pak", repos = "https://r-lib.github.io/p/pak/stable")
}
pak::pkg_install("WangLabCSU/faers")

Quick Start

The faers package provides a standardized pipeline that unifies complex pharmacovigilance workflows. For a comprehensive, step-by-step demonstration—including data acquisition and a complete Insulin case study—please refer to our detailed documentation:

The following workflow demonstrates how to perform a basic pharmacovigilance analysis for Aspirin using faers.

library(faers)

# 1. Download and Parse Data (2023 Q1-Q2)
# Note: Ensure you have enough disk space in the target directory
data <- faers(2023, c("q1", "q2"), dir = "./faers_data")

# 2. Standardization (Requires MedDRA dictionary)
data_stand <- faers_standardize(data, meddra_path = "path/to/MedDRA")

# 3. Deduplication (Requires Standardized data)
data_dedup <- faers_dedup(data_stand)

# 4.Signal Detection (Data screening for items of interest is needed, such as "aspirin".)
results <- faers_phv_signal(
    faers_filter(data_dedup, .fn = ~ drugname == "aspirin"),
    .full = data_dedup
)

Documentation

The official documentation provides comprehensive guides for both clinical researchers and bioinformaticians.

• 🌐 Official Website: The central portal for package overview and function references.
• 🚀 Full Workflow Tutorial: A complete end-to-end analysis case study (e.g., Insulin-related adverse events).
• 🧭 Getting Started: Portal: Quick-start guide and roadmap for using the faers package.

Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Report Bugs: Submit an issue if you find any calculation errors or data parsing failures.

2. Feature Requests: Have an idea for a new signal detection algorithm? Open an issue to discuss it.

3. Pull Requests:

   • Fork the project.

   • Create your Feature Branch (git checkout -b feature/AmazingFeature).

   • Commit your changes (git commit -m 'Add some AmazingFeature').

   • Push to the Branch (git push origin feature/AmazingFeature).

   • Open a Pull Request.