Context-based evidence: Pilot study to extract, repurpose, and distribute published data on OUD

Area of Science:

• * Medical Informatics
• * Data Science
• * Public Health

Background:

• * Opioid use disorder (OUD) presents a significant global health challenge with extensive research literature.
• * Synthesizing the vast body of OUD research is difficult for most individuals.
• * Existing information retrieval methods may not facilitate rapid comprehension of OUD data.

Purpose of the Study:

• * To pilot a data extraction and integration method for published OUD research.
• * To repurpose existing OUD data for enhanced comprehension and global distribution.
• * To develop a readily accessible knowledge base for opioid use disorder.

Main Methods:

• * A data-tree structure was employed to organize information from 752 selected PubMed articles on OUD.
• * Articles were screened, excluding review articles, case reports, and small sample series.
• * A hierarchical note system with parent-child relationships was utilized for data integration.

Main Results:

• * Four note types (topic, population, numeric value description, numeric value) proved sufficient for data description.
• * Data integration using shared parent notes reduced the total number of notes by 72.9%.
• * The data-tree structure allowed for flexible expansion and reduction of information nodes.

Conclusions:

• * The developed method facilitates readily understandable grouping of large datasets.
• * This strategy enables efficient extraction and accessibility of all published OUD data.
• * Improved data accessibility is expected to enhance patient outcomes and research objectives in OUD.