Step 3: Search Strategy

To identify relevant studies to include in your review, you need to perform a comprehensive literature search based on a well-designed search strategy.



The first step is to decide on which databases to search, this will depend on your research area and question. Databases differ in their coverage of journals and how articles are indexed. For preclinical research, typical databases include PubMed, Embase, and Web of Science. A librarian or an expert in bibliographic databases will be able to help you identify other potential databases and construct database-specific search terms.

On top of electronic databases, you might want to use other methods to find relevant papers such as; scanning reference lists of relevant studies (both primary studies and reviews), hand searching key journals, contacting experts in the field, and searching additional relevant internet resources. Keep a record of alternative methods used and the data collected in a structured format.


Constructing your search strategy:

Your search strategy should be constructed around your research question and ensure that your systematic review includes all relevant studies of interest. Usually the search strategy does not include the outcome(s) of interest, as these are often not reported in the title or abstract of a paper and therefore relevant studies might not be identified. For each of your research question components, you should try to consider all relevant alternative terms and synonyms, e.g. Latin forms or chemical names. If your search strategy is too narrow, you risk missing relevant studies. Too broad, however, and you may capture unnecessary studies and the screening process can become overly time consuming.

You may need to construct individual search strategies for different databases. Some databases provide manual article indexing and your search can include both free text and special dictionary terms, for example MeSH (PubMed) or Emtree (Embase) terms. Your search should be documented in such a way that it can be re-run at a later date, or by a separate group.

For preclinical research it is important to capture the studies conducted on animals. SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation) have developed open access animal filters for PubMed and Embase to assist your search. For the PubMed animal filter, which also works on Web of Science, click here. For the updated Embase animal filter click here.

The exact search strategy used is often attached or appended to a published protocol. To see examples of specific search strategies, see the SyRF Protocol Registry.


References & Resources:

Centre for Reviews and Dissemination University of York, UK

Emtree Thesaurus, Embase

MeSH Medical Subject Headings, PubMed

de Vries et al (2014) Updated version of the Embase search filter for animal studies. To copy and paste the filter click here.

Hooijmans et al (2010) Enhancing search efficiency by means of a search filter for finding all studies on animal experimentation in PubMed. For direct use as filter in PubMed click here.




  < <  Move back to Step 2   Move on to Step 4  > >