Skip to main content

Research & Publishing in Health Sciences: Advanced Searching

Learn about library resources, selecting journals, predatory publishing, and EndNote.

In order to search effectively within databases you need to familiarize yourself with these core concepts:

Boolean operators form the basis of database logic. 

  • They connect keywords and subject headings together to narrow or broaden a search.
  • The three basic boolean operators are: AND, OR, NOT

AND

AND tells a database that ALL search terms must be present in the results. Therefore, it narrows your results. An example would be artificial intelligence AND medicine AND ethics

Be aware that in some databases and search engines the AND is implied. Google Scholar automatically ANDs your search terms. 

OR

OR connects two or more similar concepts, usually synonyms, and therefore broadens your search results. OR tells a database that ANY of your search terms can be present in your search results.

So if you were to search artificial intelligence OR medicine OR ethics, you would receive results on just artificial intelligence, medicine, or ethics. You may also receive results that are about all three concepts.

NOT

Not will:

  • exclude words from your search
  • narrow your search because it tells a database to ignore a concept
  • artificial intelligence NOT medicine
Venn diagram showcasing AND as a boolean operatorVenn diagram showcasing AND as a boolean operator

Search Order

  • Databases will usually recognize AND as the primary operator and will connect concepts with AND together first.
  • If you are using a combination of AND and OR operators, enclose the words to be OR'ed together in parentheses. Example: artificial intelligence AND (medicine OR health sciences)

Truncation & Wildcards

Truncation will broaden your search because it includes various word endings and spellings. When considering truncation look for:

  • root words that have multiple endings. Example: mobility = mobilization, mobile, mobilize
  • words that are spelled differently, but mean the same thing. Example: mobilization, mobilisation
  • the database Help section. Databases use different truncation/wildcards. It is important to use the correct symbol when truncating or using wildcards. Example: *, ?, !, #

Wildcards are similar to truncation, but wildcards substitute a symbol for one letter of a word. Example: wom!m = women, woman

Keywords VS. Subject Headings

Keywords Subject Headings
are words used in natural language and describe your topic. are a predefined controlled vocabulary. They are used to describe individual content in a database.
are more flexible and can be combined together in many ways. are less flexible to search by; you must know the exact term used to describe your topic.
Databases look for keywords anywhere in the record, unlike subject terms. Subject headings only appear in subject heading or descriptor fields. In theory, if you find the exact Subject Heading that describes your topic, everything cataloged in the database should appear when searching that term.
Keywords may yield too many or too few results. If there are too many results, you can use subheadings to focus your search.
Keywords may yield many irrelevant results. Subject Heading results usually are relevant to the search topic.

MeSH

MeSH is the acronym for "Medical Subject Headings." MeSH is the authority list of the vocabulary terms used for subject analysis of biomedical literature at NLM. The MeSH controlled vocabulary is a distinctive feature of MEDLINE. It imposes uniformity and consistency to the indexing of biomedical literature.

MeSH clarifies and broadens. You can search for MeSH terms using the main search box on PubMed. Select MeSH from the drop down.

MeSH Search Box

The MeSH database always provides a definition of the term. When searching for a MeSH term, you should always ensure that the MeSH definition of the term aligns with your search strategy. In addition, you can view any subject headings associated with the MeSH term. This is the entry in the MeSH database for the term Diabetes Insipidus.

MeSH Search Result for Diabetes Insipidus

Once you find the correct term in MeSH, you can select that term by using the search box found at the right side of the page. Once you have chosen the term you wish to search, click add to search builder to place the term in the Search Builder box. Click the Search PubMed button when you are ready to search.

MeSH Search Builder

Library databases organize resources using records, and records use fields that contain specific pieces of bibliographic information. Common fields include:

  • author
  • title
  • journal title
  • abstract
  • publisher
  • date/year of publication
  • subject/descriptor

Using Fields

  • You can use fields to limit your search, which allows for more precise results.
  • To find database field types, look for drop down boxes or menus. These are often found in the Advanced Search.
  • You can combine words and fields together with boolean operators.
Venn diagram showcasing AND as a boolean operator

Different databases interpret searches differently. Phrases are a common variation among databases.

  • Some databases assume that words typed next to one another should be searched as phrases.
  • Other databases automatically add an AND between the terms of a phrase, requiring that all words be present but not necessarily adjacent to one another.
  • These two methods would return VERY different results.

Tips & Tricks

Most databases allow you to specify that adjacent words be searched as phrases. To search a phrase:

  • Use parentheses or quotes. Example: "artificial intelligence" or (artificial intelligence)
  • Check the Advanced or Guided search in a database. There is sometimes a button that you can click that indicates you want your terms searched as a phrase.

Stop Words

Stop words frequently occur and are often insignificant within the database records. Common stop words are:

  • a, an
  • the
  • in
  • of
  • on
  • if
  • be
  • are
  • into
  • which

Most databases ignore stop words. The best practice is not to use stop words in your search strategy. Instead, use keywords and field searches to eliminate the need for stop words.