Drug Evidence Intelligence: Derived Pilot Metadata for Cross-Regional Pharmacopoeial and Regulatory Evidence Comparison
Description
Copyright-safe reproducibility package for Drug Evidence Intelligence, a local research pipeline for structured extraction and comparison of pharmacopoeial, regulatory, HTA, and clinical-study evidence. This release contains derived aggregate pilot metadata for an 11-drug corpus, document-eligibility summaries, quality-comparison outcome counts, validation/audit flags, a data dictionary, and a blank manual gold-standard template. Licensed pharmacopoeial PDFs, full monograph text, long evidence excerpts, private databases, and user source documents are excluded. The included tables are pilot computational outputs and should not be interpreted as a completed clinical evidence synthesis or as evidence that one pharmacopoeia causes better clinical outcomes.
Files
Steps to reproduce
The dataset contains derived pilot outputs produced by the Drug Evidence Intelligence research pipeline. Source documents are classified by drug, source and document type. Pharmacopoeial documents are processed to extract structured quality parameters, while regulatory and clinical documents are classified according to their evidentiary role. Extracted records are normalized into a common schema and compared only when drug identity, document type, dosage form, units and other relevant comparability criteria are compatible. Supporting documents such as PSURs, RMPs and orphan-designation records are not treated as primary clinical efficacy assessments. To reproduce the computational workflow, use the Drug Evidence Intelligence source code repository together with legally obtained source documents. The repository contains the extraction pipeline, database schema, validation rules and aggregate reporting workflow. Licensed pharmacopoeial PDFs and copyrighted source documents are intentionally not included in this dataset.
Institutions
- Iran University of Science and TechnologyTehran, Tehran