Searching exercise “warm-up”
The following searching exercise is designed to walk you through the process of going from your “PICO” question to the steps of a rapid search (a few minutes) for the evidence. We will add one element of PICO at a time with the aim of getting down to around 20 potential articles. (This methed is not designed for doing a systematic review of the literature, which would be a much more extensive search).
Before you start to search for a study to answer the question we will walk through the search process with an example question.
Step 1: Create a search strategy
Scenario: A 64 year old obese male who has tried many ways to lose weight presents with a newspaper article about ‘fat-blazer’ (chitosan). He asks for your advice.
Your question in PICO format might be:
Population/problem | obese patients |
Intervention/indicator | chitosan |
Comparator | placebo |
Outcome | decrease weight |
Question: In obsese patients, does chitosan, compared to a placebo, decrease weight.
Step 2: Now convert this PICO question to a search strategy
To do this, you should do three things:
- Underline the key terms – those most specific to your question
- Number the PICO elements in order of importance from 1-4
- Think of alternate spellings, synonyms and truncations
You might end up with:
Population/problem | obes* OR overweight (2) |
Indicator (intervention, test, etc) | chitosan (1) |
Comparator | placebo (4) |
Outcome | decrease weight OR kilogram* (3) |
NOTE: ‘*’ is a truncation symbol that means further letters can be added to the word
OR finds studies containing either of the specified words/phrases, and broadens your search
AND finds studies containing both specified words/phrases, and narrows your search
1. Open your browser (eg Explorer) and go to http://www.ncbi.nlm.nih.gov/pubmed
2. Type into the search box the term we chose as (1) : ‘chitosan’. Write down the number of results you found.
3. Select Clinical Queries (left hand menu)
4. Select the Therapy category (which is the default) and search on ‘chitosan’ again. Write down the number of results you now found.
Why has this decreased? It is because of the ‘filter’ that PubMed uses to focus on clinical trials (to see the actual filter click on the ‘filter table’ on the Clinical Queries page).
5. Try adding ‘AND another stage’, ie type in chitosan AND (obes* OR overweight) – note that you need brackets around your OR search. This should reduce the number of articles even further and certainly down to a manageable number.
If we had used all terms the search may looked like this:
Search #1: | chitosan |
Search #2: | obes* OR overweight |
Search #3: | weight OR kilogram* |
Search #4: | placebo |
Search #5: | #1 AND #2 AND #3 AND #4 |
However, you might have found that the first 1 or 2 search terms were enough to narrow down the search to around 20 titles. If so you can stop!
Now apply the search strategy to a database, using the Search History button to view and combine the stages of your search:
Your own questions
Now return to the clinical questions that you developed in EBP Step 1 (Formulate an answerable question).
Use the table below to write down some search terms that you can use to get going on the search, based on your P I C O and synonyms. Number the terms in order of importance.
Question 1: | |||
Question part | Question term | Synonyms | |
---|---|---|---|
Population | ( | OR | ) AND |
Intervention or indicator | ( | OR | ) AND |
Comparator | ( | OR | ) AND |
Outcome | ( | OR | ) AND |
Remember to consider truncating words and using the * wildcard symbol, for example:
child* rather than children