Searching PubMed

Simple search is the main search bar you'll see when you first visit PubMed. It's intended to be used like a google search - simply enter your terms as they are. No need to add AND or OR operators for this kind of simple search.

This can be a helpful place to start. You can use the results to inform a more advanced search strategy, if needed.

By default, your results will be ordered by "Best Match," or relevancy, and presented in groups of ten. You can configure the display by clicking "Display Options," and selecting a different sort order (date, author, etc.), and the number of results per page.

You can also change the default "Summary" format to "Abstract," if you'd like to skim the abstracts for each reference without having to leave the results page. This also enables you to click "Find It" to access full text right from the results page.

Filters are available on the left hand side to further limit your results by date, study design, article type, language, and more. This can be particularly helpful for locating review articles!

Note - these filters rely on bibliographic data added to articles in MEDLINE. Not all references in PubMed are indexed under MEDLINE - so if you use these filters, you may be missing some things. If you need help creating a filter for a comprehensive search project, reach out to a librarian.

PubMed does not search the full text. PubMed is simply matching your search against the bibliographic information for each record (eg title, author, abstract, journal name). 

For better results, PubMed matches your search against its own subject list (MeSH, or Medical Subject Headings), and also searches your terms across common variations (eg. child to children). This is referred to as Automatic Term Mapping (ATM). Most of the times this works as you'd expect:

But sometimes there are unexpected results:

It can be useful to see how your simple search was translated. To do so, click "Advanced" under the search bar:

Under "History and Search Details," click the arrow under "Details" to see what *actually* got searched.

In this example we can see that our search terms are more or less translated appropriately. It includes "arrhythmias, cardiac" as a MeSH term, and searches cardiac arrhythmias across all other fields. It also searches the term "Drug" across all other fields. 

While there's plenty of room for improvement, this might be fine for your purposes - if it is, great! 

If you want more control, you'll want to move on to an advanced search.