GRIDINT RESEARCH · DATA HUB

Clinical trials.

Phase-three interventional trials from ClinicalTrials.gov — 987 registered studies with their enrollment, sponsor, condition, and whether results were ever posted. The registry is the U.S. National Library of Medicine's; this catalog holds its phase-three slice as filed, updated by dated edition. Every NCT number opens the trial's own page. What was not posted is counted, not guessed.

THE REPORTThe Phase-Three GateRead it →

A thousand gates, sized.

101001k10k198519901995200020052010201520202025ENROLLMENT (LOG) ↑ · START YEAR →median 224101001k10k198519901995200020052010201520202025ENROLLMENT (LOG) ↑ · START YEAR →median 224
Enrollment × start year, log scale — the mega-trials in amberEach dot is one completed phase-3 trial with a cataloged start date and enrollment (963 of the 1,000 carry both). Amber marks the top decile by enrollment; pink the top ten trials outright. Tap or hover any dot for the trial.source ClinicalTrials.gov v2 · phase-3, completed · window of 1,000 stated — method log10 enrollment · rows missing a value are absent by honestyRead the report →

Plot it on any two axes.

The shape of the records.

Records by results posted?

No results postedNo results posted: 586 of 987 (59.4%)586 · 59%Results postedResults posted: 401 of 987 (40.6%)401 · 41%987 records · counted from the records at build · source ClinicalTrials.gov API v2 — completed Phase 3 sample

SOURCE · ClinicalTrials.gov API v2 — completed Phase 3 sample · data cutoff 2026-08-26 · counted from the records

How this hub was made.

What this set is. A phase-three slice of the ClinicalTrials.gov registry — a separate pull from the report's, which is why the two do not hold the same trials; they converge to one fetch at the next edition, and until then each states its own vintage. 987 records. This pack states no retrieved date — when it was pulled is not recorded, and is not guessed at here.

Records are ClinicalTrials.gov's own (API v2), a defined sample of completed Phase 3 studies, stated as a sample and never as the registry. Every row keeps its NCT ID and links to the registry entry that owns it. Downloads are verification extracts.

Every edition, dated.

  • Edition 1 published 2026-08-27 · data cutoff 2026-08-26 · first edition

A hub changes only as a new dated edition, on the institute's word — never on a schedule. Collection continues underneath; publication is deliberate.

Verify this hub.

Verification extracts — the working records with their receipts, attribution first, pointing at the origin. We never provide data; the source is the authority, this hub is the lens.

Cite this hub

GRIDINT Research (2026). Clinical trials data hub, edition 1 (data cutoff 2026-08-26). https://gridint.ai/data/clinical-trials
@misc{gridint_clinical_trials_2026,
  title = {Clinical trials data hub, edition 1},
  author = {{GRIDINT Research}},
  year = {2026},
  note = {Data cutoff 2026-08-26; source: ClinicalTrials.gov API v2 — completed Phase 3 sample},
  url = {https://gridint.ai/data/clinical-trials}
}

In pandas

import pandas as pd
D = pd.read_json("https://gridint.ai/demo/clinical-trials/explorer-records.json")
records = pd.DataFrame(D["rows"].tolist(), columns=["nct_id", "enrollment", "results_status", "sponsor", "condition"])

The first row of that extract, as it stands: ["NCT00393952", 557, "No results posted", "SkyePharma AG", "Asthma"] — the values are the receipt, whatever a column is called.