Skip to Main Content

Evidence Based Literature Searching

  • What is the most important piece(s) of this search? (i.e. what must be in an article for it to be a relevant search result?)
  • Are there any other ways to describe this concept? (identify possibly synonyms and/or subject terms)
  • What type of resource is best to answer this question and where should I search first?

Basic Tips

  • Search each concept separately and build one at a time.
  • Think of synonyms for the words that you choose when create your PICO.
  • Try to choose a relevant database (e.g. Scopus-biomedical and life sciences).
  • Try several databases, not just one.

These tips work in most databases but check our Databases Searching Tutorials for more information.

Boolean Operators

Use Boolean Operators to broaden or narrow your search:

  • AND means both terms must be present (this will narrow your search).
  • OR means either term as well as both terms are present (this will broaden your search).
  • NOT means one term must not be present.

Word Choices

Here are some good ways to start building your words and phrases to describe each concept of your search.

  • Use Controlled Vocabulary - also known as MeSH (Medical Subject Headings) in PubMed or Subject Headings-these are terms naming descriptors in a hierarchical structure.
    • Try Focus (sometimes called Main Concept)-tells the database to retrieve only those articles in which your subject term is considered to be the primary focus of the article.
    • Try Explode-tells the database to search for your requested subject heading as well as any more specific terms that are related to your subject heading.
      Pro Tip!
      Never use all controlled vocabulary terms as this will miss articles new articles that have not been assigned terms or articles submitted to PubMedCentral (based on a grant) that in titles that PubMed does not index.
  • Use Natural Language (aka Key words) - the searcher's own words or words assigned by author's of an article.
  • Combine  Controlled Vocabulary and Natural Language to broaden your search.
    Pro Tip!
    Find some target articles and check the MeSH (PubMed) or controlled vocabulary assigned by the database for suggestions of ways to describe what is needed. Also check for keywords author's have assigned to the article (usually found under the abstract) as well as in the title or abstract. Checking these should help you build your words to search to target the best articles.

Filters

Filters can limit a large search and are offered in every database.

  • Use Limits - e.g. limiting by publication date, year, type of study and/or English language will help narrow down a search.
    Pro Tip!
    Use age group limits offered in filters with caution as this can suppress good articles.

Gold mining

Gold mining means that you are looking everywhere for “gold”, getting into the material further and pulling more information from what you have found.

  • Check the bibliography/references of a good article for more references.
  • Look for more articles written by the authors of a good article.
  • Search in targeted journals for articles matching basic key words.
Similar Articles/Related Articles

Several databases offers similar articles/related articles that match key words, terms and/or controlled vocabulary in a specific article. Below are a few examples:

  • PubMed-Simliar articles (right side of page)
  • CINAHL-Find Similar Resuls (left side of page)
  • Scopus-Related documents (right side of page)
Cited By/Cited References
Check the citations of good articles. Below are a few examples:
  • PubMed-Cited by (right side of page)

     
  • CINAHL-Go to More-Cited References

     
  • Scopus-Cited by (right side of page)

Assessing a Search Strategy

Consider these suggestions if a search is not pulling in relevant articles:

  • Re-examine your question/topic - is it too specific? Too broad?
  • Think about the search terms that you are using - consider more controlled vocabulary terms or different synonyms found in the title or  abstracts of good articles to add to the search.
  • Think about the database(s) that you are using-try another database making sure it contains relevant subjects.
  • Consult a colleague and/or librarian for suggestions about terms and databases to try - we all have different biases and experiences to draw on, might need a new perspective
  • Consider the fact that the evidence might not exist!
    This can be a hard concept to grasp in the age of technology! It might also mean that it is not available in the places/resources you are searching (proprietary, not well funded research topic, positive publication bias, etc.).
    • If this is the case the question will need to change.