Skip to Main Content

APA Formatting & Style Guide (7th Edition)

PRISMA

PRISMA stands for "Preferred Reporting Items for Systematic Reviews and Meta-Analysis." Its checklist and flow diagram template provide a framework for showing the decision-making process for systematic and systematized literature reviews.

"A PRISMA flow diagram is a visual representation of the study selection process in a systematic review or meta-analysis. PRISMA flow charts help improve transparency, methodological quality, and reporting in these types of studies, making it easier for authors, reviewers, and readers to understand the research process and assess potential biases." Read more of this summary aimed at authors like you at How to create an effective PRISMA flow diagram.

For more in-depth explanations of the full reporting standard and a copy of the PRISMA checklist with items to include in a systematic literature review paper, see the PRISMA website: https://www.prisma-statement.org/prisma-2020 

When you perform a literature search for a systematic or systematized literature review, screen the results, and select papers to write about, there are a number of things to record as you work, so that you can report them in your publication.

  • The search strategy (include in the methods and supplemental materials)
    • What to record: databases searched; subject headings and keywords used in each; search dates; and number of records returned
    • How to record: use a Word or Excel document that you can copy-paste your strategy from to recreate it.
      Advanced tipCreate a database account- Most databases provide a personal login where you can save the search strategy to rerun at your convenience.
    • Added guidance: PRISMA reporting checklist for search strategies.
  • Inclusion and Exclusion decisions (include in the PRISMA diagram)
    • What to record: number of records retrieved from each database search; number of duplicates found; and number of records and studies marked as eligible and ineligible during the two rounds of screening.
    • How to record: use folders or create tags in your citation manager. You could also create a custom field in your citation manager for recording screening decisions. For larger projects, you may prefer to use a dedicated screening tool like Covidence or Rayyan.
  • Other sources consulted (include in the extended PRISMA diagram)
    • What to include: results of snowballing, citation tracing, Google Scholar searches, society website searches, colleague-recommended items, etc.
    • What to record: names of resources; access or retrieved dates; number of items retrieved 
    • How to record: Keep a running document for websites visited and a list of the articles retrieved. This is a great thing to keep in your citation manager. Attach a copy of the document describing the website, date consulted, and any search terms used in the folder along with the retrieved items.

The PRISMA flow diagram provides a visual representation of the selection process that can be included in a manuscript to help reviewers and readers understand the decision-making process and evaluate for possible sources of bias in the review.

Figure 1 from the PRISMA 2020 Explanation and Elaboration shows the three possible inputs into the review that can be included.

  1. Previous studies (optional) - select a template for "updated" reviews if you have previously published a review on this topic
  2. The main flow from systematic database and registry searches (required) - select a template for a "new" review if is not an update
  3. Other sources (optional) - select a template that includes "other sources"

Download the appropriate Word template: https://www.prisma-statement.org/prisma-2020-flow-diagram 

Add the information about included and excluded sources to your diagram. There's a step-by-step guide here: https://guides.lib.unc.edu/prisma#s-lg-box-7601771 

You can also use this online diagram generator: https://estech.shinyapps.io/prisma_flowdiagram/

You can always choose to make your own! 

Organizing your systematic/systematized review workflow using a citation manager

This page outlines one possible way to track the numbers of references at each stage of screening using folders in a citation manager and includes notes for Zotero users. Please make an appointment to meet with a librarian to learn more about screening software, citation managers, or other possible workflows for systematized reviews. The sample data set can be found here.

1)     Create a folder (called a collection in Zotero) for each database searched. Export references from each database to a file (in most cases, a .ris file). Import each file into the relevant folder in your citation manager.

Zotero tip: Start by creating a Group rather than a Collection so that you can deduplicate over the project rather than your full library. Each set of results can go into a collection/folder within the group.

2)     Create a folder for the combined, deduplicated results and copy records from each database to the combined folder (in the image below, we have called that folder "deduplicated").

Zotero tip: CTRL-A to select all references in a folder and then drag (copies them rather than moving them) references into the combined folder. See below for a possible folder system for screening and record keeping. 

3)     Deduplicate (see the next tab above for how various tools handle deduplication)

General tip: Deduplicate after adding each database if you would like to track number removed per database. This is not needed for the standard PRISMA diagram, so recording that number is not part of this workflow. Some citation managers make it easier to delete duplicates than to keep them. If your workflow involves deleting duplicates, you can note the original number of citations in the folder name as shown above.

Zotero tip: “Duplicate Items” is an option available in the desktop version only. You will need the desktop app for deduplication. Because Zotero merges duplicates, you will still see the original number of articles from each database in each of the folders. Zotero merges duplicates, so references will continue to appear in both the original database folder and in the deduplicated/combined results folder.

4)     Screen the titles and abstracts for your results (best practice is to “blind” the view to only show titles and abstracts to remove bias from journal titles or author names)

Zotero tip: To create a blinded view, you will need to be in the desktop version where you can right click and select columns to display and show only titles in the center panel. To view the abstract, you will want to pin the abstract to the right-hand panel which effectively hides the rest of the record (right click on the abstract icon and choose “pin section”). In the right-hand panel, you can also see what folders the item appears in. You can also see where an item is located by highlighting it and holding down the ctrl (option on Mac) key.

 

 

 

 

 

5)     Move articles into the INCLUDE or EXCLUDE folders consistent with your inclusion/exclusion criteria.

General tip: Be sure to alphabetize the article list and note where you stopped if you are not able to screen all articles in one sitting (or create a temporary folder or tag to track which abstracts have been screened).

Zotero tip: Create subcollections in the EXCLUDE collection to reflect your reasons for exclusion so that you can report the numbers for each in your PRISMA diagram.

Zotero tip: If you would like to view the total number of all excluded items, toggle “Show Items from Subcollections” in the view menu.

6)     Retrieve full-text of all articles deemed relevant after title/abstract screening, See our tips on downloading articles and using ILL to request items from other libraries.

Zotero tip: PDFs can be added to Zotero! If you have a big project, add pdfs to the desktop version and go to Settings to uncheck “sync full-text content” and uncheck “sync attachment files in group libraries” to keep from hitting your cloud storage limits on the free version.

7)     Screen results based on full text. Use subcollections again to sort INCLUDED and EXCLUDED papers.

Zotero tip: Right click the reference and “add note” (or click on the note icon in the right-hand panel) to keep your notes attached to each INCLUDED article while reading. Use a standalone note to store information about the search strategies or other project-related notes not related directly to a reference.

8)     Clean up your data before writing your paper – APA style requires sentence case, and many databases will export titles in headline case or allow headline case for journals with that style. We put this at the end because it's only important to do this for articles you use to write your review.

Zotero tip: Store references in sentence case to make the citation process cleaner!

 

For basic Zotero tutorials:https://libguides.library.tmc.edu/c.php?g=1469957&p=10939111 

Zotero as research tool: https://youtube.com/playlist?list=PLTB19TB2jaeUkmZFzv_HLEYMD1fj2TeHd&si=_KwOmgQaT3MGAUU2  

After screening is complete, you will want to critically appraise the selected papers.

If you have searched in more than one database, there is a high likelihood that you will have some papers that appear in both sets.To save time during screening, it helps to deduplicate the lists against each other. Deduplication works differently in each citation manager or screening platform. Here are some links to help you with your tool of choice. To learn more, set up a meeting with a librarian.

Citation Managers

  • Zotero - Merge Duplicates - keeps both items in the original folders, but merges the data according to your preferences. If you have a large library of references and would like to auto dedupe, try adding the zoplicate plugin.
  • Mendeley - Merge Duplicates -  (Mendely can also be set up to not import items with the same doi)
  • EndNote - Bulk Deduplication - This protocol deletes the duplicates; Be sure you have noted numbers from each import before starting

Screening Platforms