A Study of Drug Recalls in the United States​

Brodie Berger

Co-Presenters: Individual Presentation

College: The Dorothy and George Hennings College of Science, Mathematics and Technology

Major: Computer Information Systems (M.S.)

Faculty Research Mentor: Ching-yu Huang

Abstract:

Every year, thousands of pharmaceutical products are pulled from the market due to safety concerns, yet the factors driving these recalls remain largely unexplored. This study explores external factors leading to drug recalls in the United States, identifying patterns in recall classifications and their distribution among pharmaceutical companies. Through data mining techniques such as ETL, outlier detection, and text mining, we analyze how recall types vary and identify key contributors to high-risk recalls.​Findings show that the most common reason for a Class 1 recalls is contamination,​ Class 2 recalls are sterility issues, while class 3 recalls deal with deviations from what is specified on packaging.​ Lastly, we uncover that few specific companies are responsible for the majority of all recalls.Analyzing these recall patterns helps improve safety regulations and can help drug manufacturing and logistic companies to know what risks they are likely to face during the manufacturing and shipping process.