A Data Pipeline for Enhancing Quality of MAUDE-Based Studies

Area of Science:

• Medical device safety
• Health informatics
• Patient safety research

Background:

• Medical device adverse event reporting systems, like MAUDE, are crucial for patient safety research.
• Current use of MAUDE data lacks a structured pipeline, impacting study reproducibility and transparency.
• Open-access MAUDE data requires robust processing for reliable research outcomes.

Purpose of the Study:

• To analyze the impact of unstructured data pipelines on MAUDE-based patient safety research.
• To propose and advocate for the implementation of an Extract, Transform, Load (ETL) pipeline for MAUDE data.
• To enhance the quality, reproducibility, and transparency of studies utilizing MAUDE reports.

Main Methods:

• Comprehensive analysis of a specific patient safety study involving endoscopic clips.
• Assessment of existing methodologies and results within the chosen MAUDE data study.
• Proposal for an ETL pipeline integrating openFDA, keyword search, and data visualization.

Main Results:

• Identified lack of structured data processing as a significant limitation in MAUDE-based research.
• Demonstrated the potential of an ETL pipeline to improve data quality and study integrity.
• Highlighted the benefits of openFDA integration and advanced search/visualization techniques.

Conclusions:

• An ETL pipeline is essential for reproducible and transparent MAUDE data research.
• ETL facilitates real-time data management and quality assurance in patient safety studies.
• Implementing structured data pipelines promotes sustainability and collaboration in medical device safety research.