Systematic and Systematic-Style Reviews

Develop a search strategy

Building a set of seed references

Before you develop your search strategy, it is important to start by doing some preliminary searching to locate resources that align with your research topic. This will help you understand the availability of existing literature and will enable you to build a set of seed references. 

These key articles can then be used to test your search strategy. Did these articles appear in your results?  If not, this may indicate a problem with your strategy. Have you:

• Searched the most appropriate databases for your topic?
• Checked for spelling mistakes or typos?
• Used a comprehensive list of search terms appropriate to your search framework (eg. PICO)?
• Used thesaurus terms/subject headings specific to each database?
• Used Boolean (and, or) appropriately?
• Added any new terms discovered while searching to all your database searches?

The importance of peer-reviewing search strategies

Submitting the search strategy used in your review for peer review helps to add veracity to your search and to eliminate bias . You can get a librarian, or someone outside your review team, to check the search strategies by following the Peer Review of Electronic Search Strategies (PRESS) checklist linked below. 

Search Strategy

The search strategy for a systematic review needs to be as comprehensive as possible in order to capture all studies relevant to the review question, and to limit bias. The strategy also needs to be transparent, rigorous, replicable and documented.

The general process in developing a search strategy is:

1. Identify synonyms and concepts related to the key concepts in your research question
2. Identify the databases and other sources you need to search 
3. Identify relevant database subject headings (e.g. see the CINAHL/MeSH subject headings in the EBSCO video below) and apply to your search strategy
4. Apply search techniques such as Boolean, Nesting, Phrase Searching, Truncation and Wildcards
5. Consider if certain concepts/keywords need to be searched in particular database fields e.g. Title/Abstract
6. Consider if certain database filters need to be applied in order to narrow the search e.g. publication dates, age, sex, article type, language etc.
7. Test, review and amend search as required

Tip: A useful starting point can be to check the search strategies in published systematic reviews for examples of how searches are structured and how they assist in keyword development.

Suggested Databases

Depending on your review question, you need to consider searching both multidisciplinary and subject-specific databases:

• Multidisciplinary databases contain a high volume of literature from a diverse range of subjects. They are helpful in locating literature on your topic from disciplines you might not have considered.
• Subject-specific databases contain in-depth coverage of literature which is highly specific to a particular field. See the library A-Z list of databases link below and refine by subject. 

Databases that have controlled vocabulary/subject headings:

Databases that do not have a controlled vocabulary or subject headings (must use keyword searching instead):

Useful resources

Finding grey literature & other sources

Grey literature is “information produced by all levels of government, academics, business and industry in electronic and print formats not controlled by commercial publishing” (GreyNet).

Read our Grey Literature Library Guide to help you to:

• identify grey literature sources 
• understand how and why to use grey literature in your research 
• apply creative search techniques for finding grey literature
• review and evaluate grey literature sources

Handsearching

Handsearching is a critical part of the systematic review process to find materials not found through traditional database searches. It is a manual process to examine and identify further relevant studies and includes:

• Perusing the pages of key journals, conferences and other sources
• Checking reference lists of identified articles and documents
• Browsing Tables of Contents of a journal or conference proceeding over a certain number of years

Read the Cochrane Training Guide to Handsearching in the link below for more information on what it is, who does it, where and what to search and what to do with the results.

Text mining tools

Text mining tools use natural language processing and machine learning techniques to analyse large volumes of text data. They can help identify relevant studies, extract data, and synthesise results. However, it's important to note that such tools should be used in conjunction with a human review process to ensure the accuracy and reliability of the findings.

Tools accessible to Southern Cross University researchers: