Skip to Main Content

How to Search the Literature (Advanced)

This guide offers an eight step approach from identifying your topic to managing your search strategy and results

Identify Search Concepts

The building blocks of a search strategy are terms. Your search strategy should contain both Subject Headings and Keywords (or text words).

Subject Headings are a standardized set of terms (called a controlled vocabulary) used to describe the content of a resource in a database. This is known as indexing.

  • For example, abstracts on cancer research can include terms like: tumour, cancer, or neoplasm. The subject heading for these terms in Ovid Medline is Neoplasms
  • By searching for Neoplasms you should retrieve all the articles where the author(s) used tumour, cancer, or neoplasm (as long as they have been properly indexed)
  • Search strategies must be customized for each database for a number of reasons, one of which is that they use different subject headings. For example:
    • Ovid Medline and PubMed use MeSH
    • Ovid Embase uses Emtree
    • CINAHL uses CINAHL Headings

Keywords are terms that appear somewhere within the resource (e.g. title, abstract, or author keywords). Searching by a keyword will retrieve resources where the author(s) used that specific term. 

  • You should include keywords in your search strategy to capture articles that have not yet been indexed, or might be indexed incorrectly

You can identify either subject headings or keywords in multiple ways.

  1. Search for a known article in a database (e.g. PubMed)
    • Click on Publication Type, MeSH terms to display the subject headings
    • Scan the title, abstract, and author keywords to identify keywords

  2. Use a concept model or map to generate keywords.
    • Write your research topic or question, along with any ideas and concepts associated with it on blank sheet of paper
    • Use themes to group your ideas, and connect related concepts using lines
      • Remember to include the following:
        • synonyms 
        • acronyms
        • brand names and generic names
        • variation in spelling (e.g. paediatric or pediatric)
  3. Search for each keyword in a database (e.g. Ovid Medline).
    • This will allow you to identify additional subject headings.
    • E.g. A search for lateral elbow yields the subject heading Tennis Elbow.

Note: not all search terms will map to an appropriate subject heading. In this case, search for it as a keyword. This option is available after the list of subject headings in most databases. 


Truncation and Wildcards

Truncation and wildcards can be applied to a keyword search in a database or search engine to broaden your results and allow you to look for variations of words.

The truncation symbol can be used in a keyword search to retrieve alternate word endings

  • Use truncation to avoid having to think of and type out all possible variations of a word
    • e.g. surg* will retrieve surgery, surgeries, surgeon, or surgical
  • Use with caution to ensure relevant words are being retrieved; in some cases, you should avoid truncating a keyword too far to the left
    • e.g. stud* will retrieve study, studies, student, students, etc.
  • Truncation symbols can vary among databases:
    • Ovid Databases (e.g. Medline, Embase, AMED, PsycINFO) truncation symbol is an asterisk (*) or dollar sign ($)
    • EBSCO Databases (e.g. CINAHL, SPORTDiscus, Social Science Index) truncation symbol is an asterisk (*)
    • PubMed truncation symbol is an asterisk (*)
    • Web of Science truncation symbol is an asterisk (*)
    • Check the help screen of other databases to find out which symbols are used

Wildcards are symbols that can optionally replace a single letter in a word

  • This is useful when dealing with variant spelling (e.g. pediatric and paediatric)
  • Wildcard symbols can vary among databases:
    • Ovid Databases wildcard is the question mark (?)
      • e.g. p?ediatric will return results with paediatric or pediatric
    • EBSCO Databases wildcard is the question mark (?) or  hash tag (#)
    • Web of Science wildcard is the dollar sign ($) or question mark (?)
    • Check the help screen of other databases to find out which symbols are used