kalamri2 / faerslib

Python tools for working with data from the FDA Adverse Event Reporting System.

FDA Adverse Event Reporting System (FAERS)

This package provides tools for working with data from the FDA Adverse Event Reporting System.

Installation

1. Install drugstandards package using PIP

sudo pip install faers

2. Installing drugstandards from source

# Download this github repository and enter the following

cd faers
sudo python setup.py install

Usage

To learn how to create the FDA Adverse Event Database check here.

Setup

# Import the module
import faers

# Initialize class and establish connection to FAERS database.
# This will create an object containing records from the UNITED STATES
# for all available years. Due to the size of the database this could take some
# time. 
f = faers.FAERS("faers.db")

# Create object using records from Japan and year 2012.
f = faers.FAERS("faers.db", countries="japan", years=2012)

# Create objects using records from United States, Japan and years 2010, 2011, 2012.
f = faers.FAERS("faers.db", countries=["japan", "united states"], years = [2010, 2011, 2012])

Utility functions

# Find associated events by drug.  Returns a list sorted by frequency.
f.associated_events("metoprolol")

# Find associated drugs by event.  Returns a list sorted by frequency.
f.associated_drugs("progressive multifocal leukoencephalopathy")

# Find drug counts.  Returns a list of drug names sorted by frequency.
f.drug_counts()

# Find event counts.  Returns a list of events sorted by frequency.
f.event_counts()

# Find events containing a phrase.  Returns a list of events containing the phrase.
f.find_like_events("fatigue")

# Find drug names by matching phrase.  Returns a list of drug names
# who's names match a given pattern/phrase.
f.find_like_drugs("pril")

Data Mining Algorithms

Proportional Reporting Ratio

# Compute Proportional Reporting Ratio (PRR) for a drug-event pair.
# Returns the PRR as well as 95% CI
f.prr("metoprolol", "nausea")

## The prr method will standardize drug terms so that following two are equivalent
# Generic name
f.prr("metformin", "nausea")

# Brand name
f.prr("glucophage", "nausea")

# Misspellings
f.prr("glucophag", "nausea")

# Compute PRR for all events given a drug name.
f.mine_prr_by_drug("natalizumab")

# Compute PRR for all drugs given an event.
f.mine_prr_by_event("progressive multifocal leukoencephalopathy")

Reporting Odds Ratio (ROR)

# Compute the Reporting Odds Ratio (ROR) for a given drug-event pair.
# Returns the ROR as well as 95% CI.
f.ror("natalizumab", "progressive multifocal leukoencephalopathy")

Multi-item Gamma Poisson Shrinker (MGPS)

# Compute the Multi-item Gamma Poisson Shrinker (MGPS) for a given drug-event pair.
f.mgps("natalizumab", "progressive multifocal leukoencephalopathy")

Questions/issues/contact

[email protected]