ACAP Library Pathfinder: Systematic Searching for a Review

The review question, thesis statement, PICO framework, and scoping searches will provide the groundwork for your search strategy.  However, It is necessary to look at all possible options to ensure you are retrieving a comprehensive set of results. 

Alternatives, Spelling & Acronyms

• Alternative spellings (e.g. counseling, counselling)
• Alternative or outdated terms (e.g. disability, handicapped, special needs)
• Abbreviations or acronyms (e.g. cognitive behaviour therapy or CBT)
• Plurals and variations (e.g. therapy, therapies, therapeutic)

Mining other sources

• Scan Gold Set articles (see Scoping Searches & Gold Sets page in this guide)
• Harvest search strategies from other systematic or literature reviews
• Use database tools (e.g., thesauri, index, subject headings) to identify controlled vocabulary terms
• Use Text Mining tools to identify recurrent words or phrases (see below)

Synonyms, Technical terms, Database terms

• Scientific or technical terms versus common or everyday language
• Consider non-English language alternatives where applicable
• Search database-specific indexing terms and search filters (see below)
• Use online and database specific thesauri

Adding Inclusion/Exclusion Criteria

• Date range
• Location (e.g. Australia)
• Participants (e.g. age, gender)
• Methodology (e.g. RCTs, qualitative, quantitative, mixed)

Controlled Vocabulary

A comprehensive search will draw on the controlled vocabulary built into the databases you are using. They are sometimes called index terms, subject headings, or thesaurus terms. Using controlled vocabulary is a powerful and precise option to add to your search strategy. Be aware that each database will have its way of indexing the terms it uses but some useful tools will help you find them. These include:

• BMI Blocks - While technically a Search Filter, it is an extremely useful tool that provides examples from medical and health bibliographic databases such as PubMed, Embase, PsycINFO, CINAHL, and others. Terms can be extracted from the blocks rather than simply copying them verbatim into your search. 
• MeSH (PubMed) - Medical Subject Headings (MeSH) thesaurus is a controlled and hierarchically-organized vocabulary produced by the National Library of Medicine.
• EbscoHost thesauri - From the search page open the SUBJECTS menu item to search individual databases' thesaurus. 
• Gold Sets -  As mentioned, you can also scan Gold Sets (the most relevant articles you've already found) from preliminary or scoping searches and other systematic review articles to harvest entire search strategies, subject headings, keywords, and index terms.

Search Filters (Hedges)

A search filter or search hedge is a ready-made search strategy developed to search complex concepts. They can be copied and pasted into database search boxes and combined with the rest of your search or edited to suit your particular search strategy. Individual terms can be harvested and added as desired. Examples include: 

• BMI Blocks - Provides examples from medical and health bibliographic databases such as PubMed, Embase, PsycINFO, CINAHL, and others.
• University of Alberta Health Sciences Filters
• ISSG - Designed to retrieve records of research using a specific study design or by topic or by some other feature of the research question
• Flinders Filters - Flinders Filters and CareSearch are developing an ever-increasing collection of subject-based search filters of national interest and international significance.
• Hedges (HIRU) -  Investigate ways to develop and hedge to improve retrieval of scientifically sound and clinically relevant study reports from large, general purpose, biomedical research bibliographic databases including MEDLINE, Embase, and PsycINFO.
• Expert Searches (OVID) - A range of search filters for use in MEDLINE and Embase - primarily focused on current health topics, diseases, and treatments.
• Methodological Search Filters - CEBMa (Center for Evidence-Based Management) - Filters include meta-analyses and systematic reviews; controlled and/or longitudinal studies.

Text Mining

Text mining software can help you identify and extract search terms by analysing text sources for recurring words and phrases. Examples of free text mining tools include:

• PubReMiner - Journal titles and MeSH terms will be displayed.
• Yale MeSH Analyzer - Enter PMIDs (PubMed IDs) and major topic terms will be displayed.
• TerMine -  Blocks of text can be mined for keywords.
• Voyant Tools - A set of tools that allows you to mine text for keywords or do text analysis. There are word and visual tools. You can paste in a block of text or add some URLs or upload documents or RIS files.
• Abstrackr - Web-based tool that uses machine learning to identify relevant studies based on your search terms. You can import references from various databases, screen them, and export them to software like Zotero.

Once a suitable research question has been determined, you can continue to use the PICO+ or similar frameworks to develop a list of comprehensive and relevant search terms. This will be an iterative process and you will find that editing this list is necessary as you work through the steps in this guide. 

Here is an example of an expanded framework based on the example question provided in the Determining the Question section of this guide. 

You might also want to use the 2dSearch app to help you organise your search terms. 

To learn about using this app take a look at the tutorials here: https://www.2dsearch.com/help.