Tip #52: Searching for PMIDs in Other Databases

What is a PMID?

A PMID, also known as the PubMed Identifier, is a unique number assigned by the National Library of Medicine to all records included in PubMed. It appears under the citation information on each record. PMIDs do not change over time or during processing and are never reused.

Searching for articles using their PMIDs can be a very efficient way to find known items or test searches (e.g., Testing for Article Inclusion in Ovid and Testing for Articles in PubMed). To search for a set of known records in PubMed, enter a string of PMIDs in the search box without the Boolean "OR" and without the [pmid] field tag. If you use [pmid], you will need to use the Boolean "OR" to combine them. These three searches return the same results:

  1. 26104772 11038025 35106283
  2. 26104772 OR 11038025 OR 35106283
  3. 26104772[pmid] OR 11038025[pmid] OR 35106283[pmid]

Below we will highlight some search tips for finding PMIDs across various databases

Too long didn't read (tldr) cheat sheet:

  • PubMed.gov - 26104772 11038025  35106283
  • MEDLINE (Ovid) - (26104772 OR 11038025  OR 35106283).ui.
  • Embase.com - (26104772 OR 11038025 OR 35106283):ui
  • Embase (Ovid) - (26104772 OR 11038025  OR 35106283).pm.
  • Web of Science - PMID=(26104772 11038025  35106283)
  • CINAHL (EBSCOhost) - PM(NLM26104772 OR NLM11038025 OR NLM35106283)
  • PsycInfo (EBSCOhost) - PM(26104772 OR 11038025  OR 35106283)
  • Scopus -   PMID(26104772 OR 11038025  OR 35106283

*The green highlighting indicates that the PMID was returned in the results. The PMIDs that are highlighted red were not returned. Continue reading to learn more about why some of the results aren't returned.

MEDLINE via Ovid 

To search for a series of PMIDs in Ovid MEDLINE, you can combine them with OR, nest them in parentheses, and add the .ui. field tag.
 
(26104772 OR 11038025  OR 35106283).ui.

Embase via Ovid

To search for a series of PMIDs in Ovid Embase, you can combine them with OR, nest them in parentheses, and add the .pm. field tag. Make sure you use .pm. and not .ui. The field tag .ui. is not a listed field in the Embase manual. While you would be correct in searching Ovid MEDLINE with .ui, it maps or defaults to the accession number tag (.an.) in Embase and returns entirely different articles. Thus a search for Using Star Wars' supporting characters to teach about psychopathology (PMID 26104772) retrieves Attitudes to HIV counseling in Bobo Dioulasso, Burkina Faso (PMID 8612016) when you search 26104772.ui.
 
Ovid Embase search 26104772.pm.

Ovid Embase search 26104772.ui.

Embase.com

The search syntax for Embase.com is very similar to Ovid MEDLINE except for the use of a colon in place of the period:

(26104772 OR 11038025 OR 35106283):ui


Embase includes many, but not all, MEDLINE records. Some PMID records may not be in Embase or may not include the PMID. For example, PMID 35106283 is not indexed for Medline and doesn't appear in our PMID search in Embase.com. The citation is in Embase.com, but without its PMID, so to find that record, you will need to search for the title.


Web of Science

You can search for PMIDs in Web of Science Core Collection, both from the Basic Search and the Advanced Search.

  1.  26104772 11038025 35106283 (select "PubMed ID" field from the drop down) in the Basic Search
  2. PMID=(26104772 11038025  35106283) in the Advanced Search Query Builder


Important to know: Web of Science may not include all MEDLINE records. In some cases, a MEDLINE record might not sync with a Web of Science, resulting in PMID results may not be retrieved even if the record is indexed in both databases.

CINAHL (EBSCO)

To search for a PMID in CINAHL, you must add the prefix "NLM" to the PMID number in addition to selecting PM Medline PMID from the Select a Field drop-down menu.

 

Important to know: Not all PubMed/MEDLINE articles in CINAHL will have PMIDs due to some differences in indexing. And not all PubMed/MEDLINE article are in CINAHL.

For example, this article is in both PubMed and CINAHL:
 
Luehring-Jones, P., Fulford, D., Palfai, T. P., Simons, J. S., & Maisto, S. A. (2024). Alcohol, Sexual Arousal, and Partner Familiarity as Predictors of Condom Negotiation: An Experience Sampling Study. AIDS and behavior, 28(3), 854–867. https://doi-org.proxy.lib.umich.edu/10.1007/s10461-023-04189-8
PMID: 37751109
 
But you won't be able to find this record by searching for the NLM37751109. 
 
You will need to search for the title.

PsycInfo (EBSCO)

Just like our searches in CINAHL, not all PubMed/MEDLINE articles in PsycInfo will have PMIDs due to differences in indexing, and not all PubMed/MEDLINE article are in PsycInfo. Starting in 2012, APA started making an effort to add more PMIDs to overlapping records, but it's still doesn't include all of them:  

"We cover some of the same journals in PsycINFO, and in April of 2012 we began adding the PMID to those records. On PsycINFO, the PMID displays in the full record view, and is a searchable field."

To search for a PMID in PsycInfo on the EBSCO platform,  type in the numbers and select the "PM PMID" field from the drop down list. You can also just type PM(26104772 OR 11038025  OR 35106283) into the search box.

Scopus.com

You can craft a search for PMIDs in Scopus from the Advanced Document search page and selecting "PubMed Identifier (PMID)" from the Field Codes > Publication list, or you can type out the following search in the Basic Search box: PMID(26104772 OR 11038025  OR 35106283)

 

As with the other non-MEDLINE databases, there is some inconsistency with whether a record will display the PMID or not. In cases where the record is included in Scopus, but without its PMID, you'll need to search for it using the title.

Comments

  1. I would love it if someone more technically inclined than me could come up with a tool that could automate the conversion of a column of PMIDs from a spreadsheet to the syntax of these various platforms (combining them with OR, etc) that i can then paste into the desired platform. It sounds like it should be simple but every time i have to do it i end up spending 10 minutes refreshing my memory on what the syntax is for that particular database, how to strip newlines, etc.

    ReplyDelete
  2. Oh - that got me to check Polyglot. Looks like a few of their translations are off!

    ReplyDelete

Post a Comment

Popular posts from this blog

Tip #1: Bulk export from Google Scholar

Tip #46: Exporting Records from Clinicaltrials.gov into EndNote

Favorite Features & Sneaky Solutions: A Database Tips Lightning Round: View the recording!