Skip to main content

Systematic reviews - searching for literature: Search strategy overview

How will I do my searching?

Searching for literature systematically is grouped into two areas:

  • The structured database search (this may be a search of several databases)
  • 'All other methods.' 

While the database search is a fundamental part of the literature search, it is useful to consider the following when undertaking your search:

"...their electronic database search only gave them 22 of their final total of 61 trials." (Knipschild cited by Greenhalgh 2006 p.119)

"...in health service management, my own team recently showed that only around a quarter of relevant, high quality papers were turned up by electronic searching." (Greenhalgh 2006 p.119)

 

Greenhalgh, T, 2006. How to read a paper: the basics of evidence based medicine. London; BMJ.

The database search

This is an outline of how you might undertake the search in a database e.g. Medline:

  1. From your topic identify key words to develop your research question, using e.g. PICO.

[Note: this is something of an iterative process and you may go back and forwards between the research topic/keywords/research question to develop all three parts]

  1. Select resources to search
  2. Identify subject headings in resources that use them
  3. Apply key words and subject headings to a database
  4. Review search based on results and amend, adding keywords
  5. Apply limiters
  6. Save search strategy
  7. Export all records to bibliographic referencing software
  8. Repeat on different databases.

'All other methods' of getting relevant articles

  • Hand search key journals
  • Forward citation search [some databases will allow you to identity articles which have subsequently cited an article you are looking at]
  • Review reference lists at the end of relevant articles
  • General search engine e.g. Google or Google Scholar
  • Current awareness services
  • Other researchers

How easy (or hard) will the search be?

Without experience it is not always so easy to tell if a search is likely to be easy or hard (you could say quick (relatively speaking) or time consuming) however it is worth remembering:

  • A focused clinical topic relating to the effect of a drug or easily defined intervention is likely to be a much easier and quicker search. This is because there are limited terms and interpretations of these terms that can be applied.
  • A focused search where words are  less easy to define e.g. about a persons 'feelings' or 'emotional response' can be difficult as there are many related terms but also a variety of contexts in which these terms can be used.
  • Individual terms e.g. 'adoption,' in the context of social work, can be problematic as the search word adopt* could be used in a huge range of contexts.

For a systematic review your research question will need to be focused, but it will also need to be sensitive. The result is that a focused search may still produce a lot of results, even though you might expect the opposite. Don't be surprised if you need to hand sift, at title and abstract level, several thousand articles.