When you run a simple search composed of just a few keywords, PubMed uses a process called automatic term mapping (ATM) to translate those keywords into MeSH terms.
Think of MeSH (Medical Subject Headings) as the index to the book that is PubMed. It is the National Library of Medicine's controlled vocabulary (thesaurus).
When new material is added to the database, it is indexed by librarians at the NLM. The librarians read each article and add appropriate information (metadata) about the article that allows searchers to find the article more easily. In addition to standard metadata (such as publication type, sex, species, and age group), the indexer also adds 10 to 15 MeSH terms that describe what the article is about. These tags are selected from a standardized set of 30,000+ MeSH terms.
By using the exact MeSH wording as part of your search, PubMed can include not just that specific term but also its synonyms, subheadings, and spelling variations. Search sensitivity is often greatly increased as a result.
The really great thing about PubMed is that it will usually be able to automatically translate the keywords you enter into their corresponding MeSH terms and include both the keyword and the MeSH term in the search string!

NOTE The most recently added references may not yet be indexed with MeSH terms.
The most effective, targeted searches employ both keywords and MeSH terms. To make sure PubMed is successfully translating your keywords into MeSH terms, you’ll need to look at the search details.
A To access the details on a search you just ran, click the Advanced link below the search box.

B At the bottom of the resulting page, below the Advanced Search Builder, is a section labeled History and Search Details.
C Toggle open the Details of a particular search to show exactly how PubMed interpreted your keywords. The image below shows how the blood draw babies query was mapped, first as a full query and then as translations of the individual components. Notice that, without any effort on your part, PubMed knew to:
The only thing it didn’t include was the singular baby; however, the inclusion of infant as a MeSH term covers that potential lapse.

Sometimes your search results are not what you expected. When this happens, you can check the details to see if something went wrong during the translation process.
Suppose we want to find articles that discuss how PPE (personal protective equipment) can prevent skin injury among hospital staff.
Our first search query—ppe skin injury prevention—produced only 22 results. A look at the details shows how our keywords, and their order, affected PubMed’s translations:

Our second query—personal protective equipment prevention skin injury—created a much more robust and successful search with 677 results. The details show that:
