Templates

Things to Consider

• What interface will you use?
• a. If you are going to use more than 1 database, then you might consider an interface that offers many databases to be searched
• 2. What databases do you want to use?
• 3. What is your research question? Can you break it down into a PICO?
• 4. Are you aiming for a wide or narrow search?
• a. I.e. will you include more or fewer words in your search?
• 5. Will you include filters or limits?
• a. Humans? Year published? English language? Types of studies?
• 6. Bonus: how have other prominent literature reviews organized their search? How does this compare to yours? You might have to look into the supplemental material of a published paper.
• Consider: You are a medical student who has been asked to research the impact of oral vs parenteral feeding in patients with acute pancreatitis. You decide to construct a literature review, and are starting with a search strategy.

What is PICO?

• P - patients with pancreatitis
• I - oral feeding
• C - parenteral feeding
• O - length of hospital stay
• T - therapy
• S - systematic reviews

• We will be limiting our search to systematic reviews (leading to a narrow search), this will allow us to scope a smaller amount of literature

Ovid Interface Example

• As pictured below, you can change the resources selected

• We have chosen the following resources in this example: EBM Reviews - Cochrane Database of Systematic reviews 2005 - February 25, 2021; Embase Classic + Embase 1947 to 2021 February 26; and Ovid MEDLINE(R) ALL 1946 to February 26, 2021
• For the purpose of this template, we will move back to one database.
• TIP → You may have to discuss with your research team which databases are best for your research question.

• TIP 2 → if you search one database, you may use the “Map Term to Subject Heading” function, which is unavailable when searching multiple databases. If you are conducting a systematic review, then you might choose to search databases separately to utilize this function. Here is what you would do below to utilize this function:

Example Limits

• We have not used any limits to our search, but below is an example of what you could use.

• TIP 3 → You can also limit your search terms to be present in the title or abstract of the article. To do this, enter your term into the search box followed by .ti,ab

Monash University

• You may need to refer to our Literature Search content for some extra help.
• TIP 4 → You can use the PubMed MeSH search to find out which terms are best for your PICO. Alternatively, you can use an interface like Ovid that offers this function when searching one database.
• After scoping Pubmed MeSH terms, we’ve determined that “enteral nutrition” would produce more results than “oral feeding” since it is a MeSH term. You may have to play around with your search in order to determine which terms are producing the best results. You can also determine which terms and subheadings are associated with your term of interest. This will be displayed in a MeSH tree (a hierarchical way of categorizing MeSH terms)
• When searching pancreatitis, we were directed to a list of terms in a MeSH tree, followed by related subheadings. This is useful because you can accept all subheadings for a wider search,or select a couple specific ones for a narrower search.

• For the purpose of this question, we have selected Diet Therapy and Therapy as our relevant subheadings. For the purpose of a systematic review, including all subheadings might be the best option.
• NOTE → We have combined these terms using the boolean operator OR.

• TIP 5 → You might limit your search term to be present in only the title and abstract. For this, you would write your term followed by .ti,ab. Our outcome was the length of hospital stay. If you are unsure how to label this term, you can use the “Term Finder” function.

Example

• You can see below how we have constructed our search. This was a narrow search for the purpose of an easy answer to our research question. We achieved this by being selective with our subheadings. For development of a systematic review, we recommend that you select many, if not all, subheadings. This is something to discuss with your research team.

• NOTE → Some databases offer a “deduplicate” function. This will ensure that you receive only one copy of a paper in your search results. This is worth trying if your search results in thousands of papers.

Review Your Results

• We have received 17 papers at the end. It is normal to end up with hundreds if not thousands. Import these citations to EndNote and begin sifting through them with your inclusion and exclusion criteria.
• Make sure to review our Navigating Endnote template next!

Finale

• After a quick scope of our literature, it seems that in patients with pancreatitis, enteral nutrition, as opposed to parenteral nutrition, is better at decreasing length of hospital stay, is associated with symptom relief and results in fewer pancreatitis-related complications.