Radiography: Search Strategy

A well-formed research question is the foundation of a successful literature search. A clear, focused question will help you identify the key concepts to search for and determine the scope of your review.

A vague question leads to an overwhelming search. A specific question keeps you focused.

Frameworks like PICO (Population, Intervention, Comparison, Outcome), SPIDER (Sample, Phenomenon of Interest, Design, Evaluation, Research type), and SPICE (Setting, Perspective, Intervention, Comparison, Evaluation) are commonly used to structure questions based on the type of review being conducted. These framework components help transform your research question into an organized search strategy by breaking down the question into searchable components

Use one of these frameworks to break your question into searchable components. 

Example using PICO:
Research Question: "What is the effect of cognitive behavioural therapy (CBT) on depression in adults?"

Once your research question is framed, the next step is to identify the key concepts. For each concept, think of synonyms, acronyms, and alternative spellings. These will be the keywords you test in your searches. For example

• Heart attack: myocardial infarction, cardiac infarction, acute coronary syndrome, heart infarction
• Student Engagement:  learner engagement, academic engagement, school engagement, classroom participation, student motivation

Start by:

• Underlining the main concepts in your question
• Brainstorming related terms or phrases
• Looking at relevant articles to see what terminology is commonly used

Tip:  Use a seed article, a relevant paper you already know, and scan its title, abstract, keywords, and references for more terms.

Create a simple table listing your concepts and their synonyms to organise your thoughts.

Controlled vocabularies are pre-defined terms used by databases to categorise articles consistently. 
Databases like PubMed, Embase, CINAHL and PsycINFO assign subject headings (e.g., MeSH terms) to articles. These help you retrieve articles even if different terminology is used by the authors.

MeSH example:
"Heart Attack":  Myocardial Infarction (MeSH)

How to Find Subject Headings:

• PubMed: Use MeSH index
• Embase: Use Emtree
• CINAHL: Use CINAHL Subject Headings
• PsychINFO: Use APA Thesaurus of Psychological Index Terms

Exploding a term includes narrower terms beneath it in the hierarchy

Tip: Start wide with “explode” and narrow later if needed.

Using controlled vocabulary helps:

• Improve the accuracy of your search
• Retrieve articles regardless of the author’s terminology

Combine these Controlled Vocabulary terms with your Keyword search for comprehensive results.

For effective search strategies, identify each key concept in your research question, then search for each concept separately by combining controlled vocabulary terms (such as MeSH or thesaurus terms) with relevant title/abstract keywords using OR operators, ensuring you capture all valuable variations before combining concepts
 

To combine keywords and controlled vocabulary effectively, use the following techniques:

• Boolean Operators: Use AND to combine different concepts, OR to include synonyms, and NOT to exclude terms
  • AND – narrows results: yoga AND blood pressure
  • OR – broadens results: heart attack OR myocardial infarction
  • NOT – excludes terms: diabetes NOT type 1 (Use NOT very sparingly)
• Phrase Searching: Use quotation marks to search for exact phrases, e.g. “cognitive behavioural therapy”, “chronic pain”
• Truncation: Use an asterisk * to find words with multiple endings (e.g. child* finds child, children, childhood)
• Wildcards: Use symbols (e.g. wom?n to find woman or women) to account for spelling variations (database dependent)
• Nesting: Use brackets to control the logic of complex searches, e.g. (child* OR adolescen*) AND (anxiety OR depression)

Try building your search incrementally and test each part.

Tip: When you're setting up your search, think of it like gathering ingredients for a recipe! For each main concept in your question, create a little collection by using OR to bring together both controlled vocabulary terms (like MeSH headings) and natural language keywords you might find in titles and abstracts. This way, you'll catch all the different ways authors might talk about the same thing before putting your concepts together

When planning your systematic search strategy, selecting the right databases is crucial for comprehensive results. Different databases index different journals and content types, so your choices should align with your research discipline and specific topic.

Start with discipline-specific databases that focus on your field, but also include key multidisciplinary databases like Web of Science and Scopus to capture relevant research published in journals outside your primary field.

Key research databases to consider:

Scopus & Web of Science: Multidisciplinary databases with extensive coverage and citation tracking capabilities
PubMed/MEDLINE: Essential for biomedical literature with free access
Embase: Comprehensive pharmaceutical and biomedical database with strong coverage of drug research
PsycINFO: Primary database for psychological and behavioral sciences literature
CINAHL: Specialized in nursing and allied health professional literature
ERIC: Focused on education research and resources
Cochrane Library: Collection of high-quality evidence databases, including systematic reviews

Browse A-Z listing of UCC databases: https://libguides.ucc.ie/az/databases

Tip: Create personal accounts to save and rerun your searches later. 

Most databases let you create a free account to:

• Save searches
• Set up email alerts
• Export results

Now it’s time to put your keywords and search techniques together to build your search.
Start with one concept and add others using Boolean logic. Refine your search based on what you find. Remember this is an iterative process.

You may need to:

• Add new keywords
• Adjust your use of controlled vocabulary
• Modify limits or filters

Keep a search log and save your search history. Use filters (e.g., publication year, language) sparingly, and only when justified.

Document your search strategy carefully so it can be reported and replicated.

Tip: Test, revise, and re-run your search.

Systematic searching is not “one and done.” Systematic Searching is Iterative Searching:You will read, revise, and refine.

Once you’ve run your searches across databases, the next step is to organise, track, and prepare your results for screening and analysis. This is a key part of keeping your review systematic and reproducible.

Export Your References
After running your searches. Export your results from each database into reference management software like:

• Zotero
• EndNote
• Mendeley

These tools help you store, organise, and cite your references with ease.

Need help choosing or using reference software? Check out our Referencing Software Guide: https://libguides.ucc.ie/referencingsoftware 

Deduplicate Your Results
When searching across multiple databases, you’ll often find duplicate references.

• Use your reference manager’s deduplication tools
• This ensures you don’t screen the same study more than once

Keep Track of Your Search Activity
Document:

• The search strategy used in each database
• The number of results retrieved
• Any changes or filters applied
• Your inclusion and exclusion criteria

Tip: Use a search log or audit trail (Excel or Word) to stay organised:

This Excel template is very useful:  A spreadsheet designed to be used for recording and summarising search strategiesis  You can download and adapt the EPOC Search Audit Worksheet: 

• Paul Miller, on behalf of Effective Practice and Organisation of Care (EPOC). Search Audit Template Excel Spreadsheet. Zenodo, 15 July 2021., https://doi.org/10.5281/ZENODO.5106379.
•  

Use Screening Tools
For collaborative screening and study selection, try:

• Rayyan: A free web-based tool for screening and tagging
• Covidence: Covidence: A specialized platform designed for systematic reviews that facilitates screening, data extraction, and consensus building. Note that UCC Library does not have an institutional subscription to Covidence, though some individual departments maintain their own subscriptions. Check with your department about access options.

Both tools allow you to:

• Blind-screen studies
• Apply inclusion/exclusion criteria
• Tag and organise decisions

Save and Back Up Everything

• Save your search history in each database (most allow you to create a personal account and save searches)
• Keep a secure backup of all exported files, logs, and notes

As said above, Systematic searching is not “one and done.” You will read, revise, and refine. Systematic searching is iterative.

Keeping well-managed records ensures transparency, accuracy, and efficiency throughout your review.