PubMed Scraper – Articles, Abstracts, Authors, Affiliations & Emails

Search PubMed (36+ million biomedical citations) and export clean rows: title, abstract, authors with affiliations, ORCID and email addresses, journal, publication date, DOI, PMC free-full-text link, MeSH terms, keywords, publication type and grants. Use full PubMed search syntax, or look up PMIDs in bulk. A monitoring mode returns only articles you have not received before, so a scheduled run becomes a literature alert.

• Official NCBI E-utilities API. No scraping of web pages, no browser, no login. An NCBI API key is optional (it only raises the speed limit).
• Gets past the 9,999-result limit. E-utilities stop at 9,999 records per search; this Actor continues year by year.
• $1 per 1,000 articles. Not-found PMIDs, duplicates and skipped rows are free. Start fee $0.00005.

What can I use it for?

• Literature reviews and meta-analyses: export every article for a query with abstracts and MeSH terms to Excel, CSV or JSON.
• Research lead lists / KOL mapping: find the authors publishing on a topic, with affiliations and the emails they put in their papers.
• Literature alerts: schedule onlyNew daily or weekly per drug, disease, competitor or author.
• AI/RAG datasets: abstracts plus structured metadata, ready for embeddings.
• Citation lists: a ready-made citation string for each article.

Input

{
  "queries": ["semaglutide obesity", "Smith J[Author] AND 2025[dp]"],
  "maxResultsPerQuery": 100,
  "sort": "newest",
  "dateFrom": "2024",
  "onlyWithAbstract": true
}

Field	What it does
queries	PubMed searches; field tags work: [Author], [Journal], [MeSH], [Affiliation], [pt], [dp], boolean AND/OR/NOT
maxResultsPerQuery	Articles per query (0 = all)
sort	Best match, newest, oldest, first author, journal
dateFrom, dateTo	Publication date range (YYYY, YYYY/MM or YYYY/MM/DD)
onlyFreeFullText, onlyWithAbstract	Filters
pmids, bulkPmids	Fetch specific PMIDs
onlyNew + stateKey	Monitoring: output only PMIDs not returned in earlier runs
ncbiApiKey	Optional free NCBI key (10 instead of 3 requests per second)

Output

{
  "pmid": "42850698",
  "title": "Weight-Loss Outcomes After Sleeve Gastrectomy Versus Tirzepatide in a Minority Cohort With Severe Obesity...",
  "abstract": "While GLP-1 and GIP receptor agonists like semaglutide and tirzepatide have revolutionized ...",
  "authors": [{"name": "Emiliano G Manueli Laos", "orcid": null, "affiliations": ["..."], "emails": []}],
  "firstAuthor": "Emiliano G Manueli Laos",
  "lastAuthor": "...",
  "affiliations": ["..."],
  "emails": ["..."],
  "journal": "Clinical obesity",
  "journalAbbrev": "Clin Obes",
  "pubDate": "2026-12",
  "year": 2026,
  "doi": "10.1111/cob.70116",
  "pmcid": null,
  "isFreeFullText": false,
  "publicationTypes": ["Journal Article"],
  "meshTerms": ["Humans", "Tirzepatide", "Weight Loss", "..."],
  "keywords": ["..."],
  "grants": [],
  "citation": "Emiliano G Manueli Laos, ... Clin Obes. 2026;16(6):e70116. doi:10.1111/cob.70116. PMID: 42850698.",
  "pubmedUrl": "https://pubmed.ncbi.nlm.nih.gov/42850698/"
}

The Articles and Authors & emails tables in the Output tab show the main columns. In a test search about 4 in 10 recent articles carried at least one author email.

Pricing

Pay per event: $0.001 per article ($1 per 1,000) + $0.00005 per run.

Actor (Apify Store, Oct 2026)	Price per 1,000 articles
PubMed Scraper (this Actor)	$1
ryanclinton/pubmed-research-search	$2
easyapi/pubmed-search-scraper	$3 + $0.09 per run
scrapestorm/pubmed-articles-scraper	$10

FAQ

Is this allowed? The Actor uses NCBI's public E-utilities API and follows NCBI's usage policy (tool and email identification, at most 3 requests per second without a key). PubMed citations are public, but some abstracts are copyrighted by their publishers. You are responsible for how you reuse the text.

Why are some emails missing? PubMed only has the emails authors put in their affiliation line, usually the corresponding author.

How does onlyNew work? The PMIDs returned are stored in the named key-value store pubmed-search-state under your stateKey. Later runs skip them (free), so you only pay for articles you have not seen.

Can I get full text? Free full text is on PubMed Central; use pmcUrl or filter with onlyFreeFullText.

Use with AI agents / Apify MCP

Add the Actor to the Apify MCP server (mcp.apify.com) and ask: "find 2025 randomized trials on GLP-1 and alcohol use, with abstracts". Rows are flat JSON that an LLM can summarize directly.

Related Actors

• PDF Text Extractor – text from PDF papers.
• Website to Markdown – clean Markdown for RAG.