Pubmed Data Extractor

PubMed Data Extractor is a fast scraper that turns PubMed search results into structured, exportable data. Give it one or more keywords and it collects the citation details you would otherwise copy by hand: titles, authors, journals, publication years, DOIs, PMIDs, and abstract snippets. It is built for research, competitive content analysis, and SEO work that needs real data from published medical literature.

PubMed indexes more than 38 million citations from MEDLINE, life science journals, and online books. Reading through that by hand does not scale. This actor does the searching, reading, and paging for you, and hands back a clean dataset you can drop into a spreadsheet, a database, or your own app.

Why use it

• Save hours of manual work. One run replaces dozens of copy-and-paste passes across search pages.
• Get clean, structured fields. Journal name, year, and DOI are parsed out of the raw citation for you, so the data is ready to sort and filter.
• Search many topics at once. Pass a list of terms and each one is searched separately in the same job.
• Own your data. Export to CSV, JSON, or Excel, or pull it through the Apify API into your own pipeline.

Who it is for

• SEO and content teams building articles around trusted medical sources
• Medical writers and researchers gathering citations for reviews
• Data teams tracking new publications on a condition, drug, gene, or method
• Anyone who needs a repeatable, structured feed of PubMed results

Features

• Search PubMed by keyword, phrase, author, or query
• Run multiple search terms in a single job, each with its own result limit
• Collect title, authors, journal, year, DOI, PMID, abstract snippet, and article URL
• Automatic pagination across result pages
• Deduplication by PMID within each term
• Reports the total number of matches PubMed found for every term
• Export to CSV, JSON, Excel, or read through the API

How it works

1. You enter one or more search terms, the same way you would type them on the PubMed website.
2. For each term, the actor runs the search and reads every result on the page.
3. It pages through the listing until it reaches the number of items you asked for.
4. Each article is parsed into a clean row and saved to the dataset.

Input

Field	Type	Required	Description
searchTerms	array of strings	Yes	One or more keywords, phrases, authors, or queries. Each term is searched separately.
maxItems	integer	No	Maximum results to collect for each search term. Default 10, maximum 1000.
requestTimeoutSecs	integer	No	Per-request timeout in seconds. Default 30.

Example input

{
    "searchTerms": ["Rheumatoid arthritis", "CRISPR gene editing"],
    "maxItems": 25
}

This searches both terms and returns up to 25 results for each, so up to 50 items total.

Output

Each result is saved as one dataset item.

{
    "pmid": "38354510",
    "articleTitle": "Rheumatoid arthritis autoantibodies benefit from inflammation temperatures.",
    "authors": "Singh PK, Stan RC.",
    "journalCitation": "Int Immunopharmacol. 2024 Mar 10;129:111690. doi: 10.1016/j.intimp.2024.111690.",
    "journalName": "Int Immunopharmacol",
    "publicationYear": "2024",
    "doi": "10.1016/j.intimp.2024.111690",
    "snippet": "BACKGROUND: Symptoms of rheumatoid arthritis are associated with local inflammation and may include low-grade fever...",
    "url": "https://pubmed.ncbi.nlm.nih.gov/38354510/",
    "searchTerm": "Rheumatoid arthritis",
    "resultPosition": 3,
    "totalResults": 187602,
    "scrapedAt": "2026-07-06T12:00:00+00:00"
}

Output fields

Field	Type	Description
pmid	string	PubMed identifier for the article.
articleTitle	string	Article title.
authors	string	Author list as shown on the result.
journalCitation	string	Full citation string, including volume, pages, and DOI.
journalName	string	Journal name parsed from the citation.
publicationYear	string	Publication year parsed from the citation.
doi	string	DOI parsed from the citation, when present.
snippet	string	Abstract snippet shown under the result, when present.
url	string	Direct link to the article on PubMed.
searchTerm	string	The term that produced this result.
resultPosition	integer	Position of the result on its page.
totalResults	integer	Total matches PubMed reported for the term.
scrapedAt	string	UTC timestamp of the run.

Use cases

• Content research. Pull the latest papers on a topic and cite them in blog posts, guides, or landing pages.
• Literature tracking. Run the same terms on a schedule to watch for new publications.
• Competitive analysis. See which journals and authors dominate a subject before you write about it.
• Dataset building. Collect thousands of citations across many terms to feed a model, an app, or an internal knowledge base.

Performance and limits

• Page size adapts to your maxItems, so small runs use a single request per term.
• Results are deduplicated by PMID within each term.
• Broad terms can match hundreds of thousands of records. The totalResults field shows how many exist; use maxItems to keep runs focused and fast.
• Results reflect what PubMed returns at run time, in the same order the website shows them.

Tips

• Use quotation marks for exact phrases, for example "gene therapy".
• PubMed field tags work too, for example asthma[Title] or Smith J[Author].
• Combine terms with AND, OR, and NOT for more precise searches.

FAQ

Can I search several topics in one run? Yes. Add each topic to searchTerms and the actor searches them one after another, each with its own limit.

How many results can I get per term? Up to 1000 per term. The default is 10.

What format is the output? A structured dataset you can export as CSV, JSON, or Excel, or read through the Apify API.

Does it return abstracts? It returns the abstract snippet shown on the search result. Full abstracts live on each article page, linked in the url field.

Notes

This actor collects publicly available bibliographic data from PubMed search results. PubMed is a service of the U.S. National Library of Medicine. Use the data in line with PubMed's terms and give proper attribution to the original publications.