PubMed & ClinicalTrials.gov Research Data

Search and extract medical research data from PubMed (35M+ biomedical articles) and ClinicalTrials.gov (500K+ clinical studies). Two of the world's most important biomedical databases in one Actor.

Why use this Actor?

• Official NIH data — direct from the National Library of Medicine and National Institutes of Health
• No API key required — both APIs are free and open
• No proxy needed — accessible worldwide
• 2 databases in one — PubMed articles + ClinicalTrials.gov studies
• Rich metadata — authors, journals, MeSH terms, DOIs, trial phases, sponsors, outcomes
• Abstracts included — full abstract text for PubMed articles

How much does it cost?

This Actor uses pay-per-event pricing. You are charged per result item returned.

No proxy costs. No API key costs.

Two data sources

PubMed

Over 35 million citations and abstracts of biomedical literature. Covers medicine, nursing, dentistry, veterinary medicine, health care systems, and preclinical sciences.

Returns: article title, authors, journal, publication date, DOI, abstract, publication type.

ClinicalTrials.gov

Over 500,000 clinical studies from 228 countries. The world's largest database of clinical trials.

Returns: NCT ID, title, status, conditions, interventions, phases, enrollment, sponsor, outcomes, eligibility criteria.

Input parameters

Parameter	Required	Default	Description
source	Yes	pubmed	pubmed or clinical_trials
query	Yes	-	Search terms
dateFrom	No	-	Start date
dateTo	No	-	End date
studyStatus	No	all	ClinicalTrials: filter by status
includeAbstract	No	true	PubMed: include abstract text
maxResults	No	100	Maximum results

Example: PubMed output

{
    "pmid": "38123456",
    "title": "Advances in Cancer Immunotherapy: A Comprehensive Review",
    "authors": ["Smith J", "Johnson K", "Williams R"],
    "firstAuthor": "Smith J",
    "journal": "Nature Reviews Cancer",
    "pubDate": "2026 Jan",
    "doi": "10.1038/s41568-026-0001-2",
    "abstract": "Cancer immunotherapy has transformed oncological treatment...",
    "pubType": ["Review"]
}

Example: ClinicalTrials.gov output

{
    "nctId": "NCT06123456",
    "title": "Phase 3 Study of Drug X in Advanced Melanoma",
    "status": "RECRUITING",
    "conditions": ["Melanoma"],
    "interventions": ["Drug X", "Placebo"],
    "phases": ["PHASE3"],
    "enrollment": 450,
    "sponsor": "Pharma Corp",
    "countries": ["United States", "Germany", "Japan"],
    "primaryOutcomes": ["Overall Survival"]
}

Use cases

• Literature review — search PubMed for specific topics, drugs, or diseases
• Competitive intelligence — track clinical trials by competitor, drug, or indication
• Drug development — monitor clinical trial landscape for specific conditions
• Academic research — build datasets for meta-analyses and systematic reviews
• Pharmacovigilance — track safety-related publications and trial results
• Market research — analyze pharmaceutical R&D pipelines

FAQ

Q: Can I search for clinical trials by drug name? A: Yes. Set source to clinical_trials and use the drug name as your query. You can also filter by study status (e.g., RECRUITING) to find active trials.

Q: How do I find systematic reviews or meta-analyses? A: Search PubMed with terms like "systematic review" or "meta-analysis" in your query. PubMed's search engine recognizes publication type filters.

Q: What's included in the abstract? A: Full abstract text when available. Set includeAbstract: true (default). Some older articles and conference abstracts may not have full text abstracts.

Q: How far back does PubMed data go? A: PubMed contains citations dating back to the 1950s, with more complete coverage from the 1970s onwards. Currently over 35 million records.

Q: Can I search by MeSH terms? A: Yes. Include MeSH terms in your query — PubMed's search engine automatically matches against MeSH headings. For example, searching "neoplasms" will match articles tagged with the Neoplasms MeSH term.

Q: What clinical trial statuses are available? A: Available statuses include: RECRUITING, ACTIVE_NOT_RECRUITING, COMPLETED, TERMINATED, WITHDRAWN, NOT_YET_RECRUITING, SUSPENDED, and ENROLLING_BY_INVITATION.

Integration with Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")

# Search PubMed for recent cancer immunotherapy reviews
run = client.actor("aligned_kettledrum/pubmed-clinical").call(
    run_input={
        "source": "pubmed",
        "query": "cancer immunotherapy review",
        "dateFrom": "2025-01-01",
        "includeAbstract": True,
        "maxResults": 100,
    }
)

# Analyze results
import pandas as pd
items = list(client.dataset(run["defaultDatasetId"]).iterate_items())
df = pd.DataFrame(items)

# Publications per journal
journal_counts = df["journal"].value_counts().head(10)
print("Top journals:")
print(journal_counts)

# Extract all unique authors
all_authors = [a for authors in df["authors"] for a in authors]
print(f"\nTotal unique authors: {len(set(all_authors))}")

Technical details

• PubMed: Uses NCBI E-utilities (esearch, esummary, efetch). Rate limited to 3 requests/second without API key.
• ClinicalTrials.gov: Uses the v2 API released in 2023. Returns structured JSON data.
• Both APIs are maintained by the U.S. National Institutes of Health.

MCP Server

Need PubMed data inside your AI agent? Use the PubMed MCP Server — same data, native MCP integration for Claude, ChatGPT, and other LLM frameworks.