# Faers

> faers

- Skill: `biomate-ai/faers` (Agent Skill)
- Install (CLI): `npx skillmds@latest add biomate-ai/faers`
- Raw SKILL.md: https://api.skillmd.com/api/skills/biomate-ai/faers/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: biomate-ai (https://skillmd.com/u/biomate-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/biomate-ai/faers

---


# faers

## Workflows

### Standard Workflow

```r
library(faers)

# 1. Check metadata of FAERS
faers_meta(internal = TRUE)

# 2. Download and parse quarterly data files
data2 <- faers(
  c(2004, 2017), 
  c("q1", "q2"), 
  dir = system.file("extdata", package = "faers"), 
  compress_dir = tempdir()
)

# 3. Retrieve specific field data (e.g., demographic data)
demo_data <- faers_get(data2, "demo")

# 4. Load standardized data and retrieve MedDRA hierarchy
data_std <- readRDS(system.file("extdata", "standardized_data.rds", package = "faers"))
meddra_hier <- faers_meddra(data_std, use = "hierarchy")
```
**Input/Output Note:** Inputs are year, quarter, and directory paths for FAERS raw data; outputs are standardized FAERS data tables and MedDRA hierarchy mappings.

## When to Use
- Accessing and parsing the FDA Adverse Event Reporting System (FAERS) quarterly data files.
- Retrieving specific demographic, drug, or reaction fields using `faers_get`.
- Standardizing drug and reaction terms to MedDRA terminology using `faers_standardize` and `faers_meddra`.

## When NOT to Use
- For analyzing clinical trial adverse events that are not part of the FAERS public database.
- For real-time adverse event tracking (FAERS data is released quarterly).

## Data Requirements
- Raw FAERS quarterly data files (ASCII or XML format).
- MedDRA terminology files for standardization.

## Key Parameters
- **internal** (`FALSE`): Logical indicating whether to use internal cached FAERS metadata in `faers_meta`.
- **dir**: Directory path where FAERS data files are stored or downloaded.
- **compress_dir**: Temporary directory for uncompressing downloaded files.
- **use** (`"hierarchy"`): Component of MedDRA to retrieve (e.g., `"hierarchy"` or `"SMQ"`).

## Best Practices
- Check FAERS metadata using `faers_meta(internal = TRUE)` to verify available quarters and file sizes before downloading.
- Use `faers_combine` to merge multiple quarters of parsed FAERS data into a single object.
- Standardize drug and reaction terms using `faers_standardize` with a valid MedDRA path to ensure consistent terminology.

## Common Pitfalls
- Attempting to run standardization without providing a valid path to uncompressed MedDRA data. Ensure `meddra_path` is correctly specified.

## Alternatives
- `openfda`: For querying the OpenFDA API directly instead of parsing raw quarterly files.

## Citations
- FDA Adverse Event Reporting System (FAERS) documentation.

## References
- Homepage: bioconductor.org/packages/faers
- Vignette: https://bioconductor.org/packages/release/bioc/vignettes/faers/inst/doc/FAERS-Pharmacovigilance.html