Developing a good systematic search strategy is key to finding the best available evidence efficiently.
The search must be sensitive enough to pick up the most relevant papers to our question, and specific enough that we can find the answer quickly. A poorly designed search strategy with insufficient search terms, or errors in the search strategy, may not pick up all the relevant literature, and a strategy that is too broad and picks up superfluous articles runs the risk of you missing the relevant articles in amongst the irrelevant ones, as well as taking up valuable time screening titles and abstracts.
Do you remember what MeSH is? Jump to Library databases: Medline for a refresher.
Combining MeSH terms with appropriate keywords and synonyms helps to pick up as many relevant papers as possible – this makes the search sensitive. Good ways to come up with search synonyms are:
Using truncation symbols and wildcards with search terms helps to make the search even more sensitive, by picking up variations in spelling and word endings.
Potential pitfalls: failing to identify appropriate MeSH terms; spelling errors in search terms; inappropriate or truncation (too far to the left of the word – picking up irrelevant terms or too far to the right/not used at all – failing to pick up relevant papers using variation of the root word)
By combining search terms for the different elements or concepts with AND, the search then becomes more specific. We can make the search even more specific by limiting our search by date, population or study design.
Sometimes it isn’t practical to use all the elements of our research question in our search strategy – we might find it makes the search too specific, and we find too few papers.
Potential pitfalls: Incorrect use of Boolean operators; incorrect search logic – this can be prevented by separating your searches on different lines for each PICO element – or even a separate line for MeSH terms and keywords for each PICO.
Separating searches for each term onto different lines can also help with refining the search – you can then easily alter or remove search terms that are affecting your search sensitivity or specificity.
For more on searching databases and examples, see our Search Tips section.
Creating a search strategy that has each PICO element separated onto a different line – and preferably separating out MeSH terms and keywords as well – helps in refining your search. If any search terms seem to give very few results, you may need to refine it further with truncation or wildcards or combining with additional synonyms using OR. Combine all PICO elements and see whether your search strategy gives you a manageable number of results. If there are still too many results to easily skim, then perhaps refine by date of publication (last 5 years or 10 years) or study types (systematic review or RCTs for intervention).
If you have too few results, then try removing less specific search terms – in this case, perhaps remove the search terms related to being female. Then perhaps try removing the outcome terms, and instead include these elements into your eligibility criteria, for application during the paper screening process.
Use of truncation symbols and wild cards can help to pick up variations of your search terms, especially US versus UK spellings, and ending variations eg prostat* will pick up prostate AND prostatic (and also prostatectomy).
Syntax: Which advanced search syntax (e.g. truncation and wildcards) will you use?
Limits: Which limits will you apply to your search results? (e.g. age, gender of patients; language; publication date; geography)
Study design:
Consider whether you want to limit by study design, by date of publication, by age range, by language and use the search engine delimiters. Be sure to have a good rationale for these limits.
Screening is the process of identifying which studies from your literature search(es) will be included in your review. Each individual article must be assessed to see if it meets the eligibility criteria you've considered.
The screening process involves:
For further details and tools to help see our Systematic and systematic-like review guide - Screening.
There are a number of options for finding good quality evidence. For systematic reviews and RCTs, the Cochrane Library. For primary research, journal databases such as MEDLINE (using the Ovid or PubMed platforms) for a medical-focussed database or Emcare (Ovid) or CINAHL (EBSCOhost) if your question overlaps with nursing and allied health, or PsycINFO (Ovid) if your topic is within the discipline of psychology.
MEDLINE is a commonly-used medical journal database, which is why it is recommended for this assignment.
Google Scholar can be useful when you know the article you are looking for, or doing a quick scoping search for background information, but doesn’t have the powerful filters and tools that specialised databases do.
To view and access the library databases go to the A-Z Databases: Medical Sciences page. Use this guide to find out how to effective search the databases:
Use the PICO elements to brainstorm keywords and to look up MeSH terms.
Key concept: Identify the key components of your clinical question and break them up using PICO
Synonyms: What other search terms could be used to describe your PICO concepts? Can often be found by scanning titles and abstracts in your initial reading around the topic.
Subject headings: Which subject headings(MeSH terms) could be used to describe each concept?
| Population | Intervention | Comparison | Outcome | |
|---|---|---|---|---|
|
Key concept
|
Iron deficiency anaemia Non-pregnant Woman Adult |
Iron polymaltose |
Ferrous sulfate |
Ferritin Hb Side effects |
| Synonyms |
Iron deficiency anemia IDA Women Female |
Iron polymaltose complex IPC Maltofer Polysaccharide-iron complex PIC |
Iron polymaltose complex IPC Maltofer Polysaccharide-iron complex PIC |
Adverse effects |
| Subject headings |
Anemia, Iron-deficiency |
Ferrous compounds |
“Drug-Related Side Effects and Adverse Reactions” |
A quick way to look up The Medical Subject Headings (MeSH) is through National Library of Medicine thesaurus database or use the "Term Finder" in Medline.
Charles Sturt University acknowledges the traditional custodians of the lands on which its campuses are located, paying respect to Elders, both past and present, and extend that respect to all First Nations Peoples.
Charles Sturt University is an Australian University, TEQSA Provider Identification: PRV12018. CRICOS Provider: 00005F.
