Dataset
clinicaltrials-database
A queryable DuckDB mirror of the AACT flat-file export of ClinicalTrials.gov, packaged via the clinicaltrials-database project and covering every registered trial across 48 tables totaling roughly 58 million rows.
Data preview
Dataset card
The clinicaltrials-database is a DuckDB-formatted mirror of the AACT (Aggregate Analysis of ClinicalTrials.gov) flat files, packaging the full ClinicalTrials.gov registry into a single queryable file for offline analysis. It is published by Nason and tagged as a 10M–100M-row tabular dataset covering clinical study records.
What this dataset contains
The dataset mirrors the AACT relational export of ClinicalTrials.gov, which is maintained by the Clinical Trials Transformation Initiative and updated daily. Because AACT is the canonical structured export of every registered study, the file includes protocol-level records, site and sponsor information, interventions, outcomes, and eligibility criteria. Coverage is effectively global wherever sponsors register trials in the U.S. registry.
Structure and fields
Records are organized in the standard AACT relational schema, including tables such as studies, sponsors, facilities, interventions, conditions, and outcomes, with categorical fields like overall_status and phase stored as UPPER_SNAKE strings (e.g. 'COMPLETED', 'PHASE3', 'DRUG'). The DuckDB build attaches the database read-only and supports HTTP range requests, so analysts can SELECT from ct.studies or join across tables without downloading the whole file. A header-only phantom table exists in the source files and is intentionally not loaded.
How analysts use it
Analysts typically query the file with DuckDB to track trial pipelines by sponsor, condition, or phase, and to monitor reporting timeliness against FDAAA anticipated-versus-actual dates. For investment work, the dataset can be used to size addressable patient populations, gauge sponsor R&D productivity, and benchmark biotech catalysts by counting active studies in a given indication. The public-domain U.S. government status of the underlying data makes it suitable for redistribution in internal research.
Documentation & references 1
Related datasets
Browse moreParsed eligibility criteria from 13,229 ClinicalTrials.gov studies representing the candidate pool gathered by two first-stage retrievers during TREC Clinical Trials 2021–2023, formatted as typed entity-relation graphs for reranking models.
The clinical-trials-patient-graphs dataset is a small tabular collection of patient-level clinical records distributed in Parquet format by the 2001jdev publisher. It organizes entities and relations extracted from trial data into structured rows spanning 2021 through 2023.
This dataset is a parsed snapshot of ClinicalTrials.gov records, distributed in Parquet format on the Hugging Face Hub under an unspecified license. It contains roughly 52,000 trial entries drawn from public registry filings.
clinical-trials-trec-qrels is a tabular relevance-judgment file distributed by the 2001jdev user on the Hugging Face Hub. It maps clinical-trial topic identifiers to NCT registry IDs with graded relevance scores.
The clinical-trials-trec-topics dataset is a small tabular collection of clinical-trial topic records distributed in Parquet format on the Hugging Face Hub. It appears to be derived from TREC clinical-trial retrieval topics, with rows partitioned by year.
SEC EDGAR 10-K & 10-Q financial statements for 2,600+ NASDAQ companies.
Discussion
Questions or observations about this dataset? Join the conversation on X.
No comments yet. Start the conversation on X.