FAERS-Database

OpenSource FAERS Database Cleansing, Merging and Analysis

Technologies used:

1- Postgresql version 4.21

2- Python

Steps to understand this project:

1- Business Understanding of needed research question and DB information

2- Data Understanding of FAERS Database

3- Importing since 2012 Q1 to 2020 Q2

4- Merging Tables

5- Removing Full Row Duplicates

6- Data Quality Issues ex. Age

7- Data Cleansing ex. Gender

8- Data Engineering ex. Age_cod and age_grp

9- Python to Extract meddra codes files into txt file

10- Importing Meddra Codes and names

11- Importing Soc codes and names

12- Removing Duplicates

13- Calculating 2 by 2 table information to get: A,B,C,D

14- Caulcation of ROR, PRR, IC, EBGM and CI95% for each of them

15- Demographic Analysis

16- Visualizations

The data of the FAERS belongs entirely to the FDA and can be accessed as open source from this website:

https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html
the ASCII files contain the CSV data that we used delimited by a $ sign

The data of the MedDRA is not open source, however you can learn more about it from these website:

1- MedDRA official website: https://www.meddra.org/ which has very easy presentation for its explanation found at: https://meddra.org/sites/default/files/page/documents_insert/meddra_-_terminologies_coding.pdf

2- MedDRA dictionary in oracle: https://docs.oracle.com/health-sciences/argus-suite-821/AADMN/dictionaries.htm#AADMN694

3- The online version of MedDRA: http://bioportal.bioontology.org/ontologies/MEDDRA/?p=classes&conceptid=root

Because of MedDRA not being open source so we can only share that we used MedDRA version 22.1.

Acknowledgements

Please cite this github and the following paper(s) if you use it in your work:

Authors:

Marwa Eshra (https://eg.linkedin.com/in/marwa-eshra) (ORC ID: https://orcid.org/0000-0001-5600-1166)

Tariq Saeed (https://eg.linkedin.com/in/tariq-saeed-a5279613) (ORC ID: https://orcid.org/0000-0001-9774-0483)