PRISMA Flow Diagram Generator
Enter the counts from your search and screening — records identified, duplicates removed, records screened and excluded, reports assessed and excluded with reasons — and get the PRISMA 2020 flow diagram for a new systematic review, laid out like the official template. Steps you leave blank are worked out from the step before, and numbers that do not add up are flagged.
Fill in the counts you have. A blank step is worked out from the step before it, and a number that does not add up is flagged below the form. Blank removals count as 0.
Identification
Screening
Included
Other methods
The numbers add up from identification to inclusion.
Show the counts as a table
| Box | n |
|---|---|
| Databases | 2,257 |
| — MEDLINE (Ovid) | 812 |
| — Embase | 1,054 |
| — APA PsycInfo | 391 |
| Registers | 48 |
| — ClinicalTrials.gov | 31 |
| — WHO ICTRP | 17 |
| Duplicate records removed | 612 |
| Records marked as ineligible by automation tools | 87 |
| Records removed for other reasons | 0 |
| Records screened | 1,606 |
| Records excluded | 1,498 |
| Reports sought for retrieval | 108 |
| Reports not retrieved | 6 |
| Reports assessed for eligibility | 102 |
| — Wrong population | 31 |
| — Wrong intervention | 22 |
| — Wrong outcome | 17 |
| — Not a randomized trial | 11 |
| Studies included in review | 19 |
| Reports of included studies | 24 |
| Other methods: Websites | 9 |
| Other methods: Organisations | 3 |
| Other methods: Citation searching | 14 |
| Other methods: Reports sought for retrieval | 12 |
| Other methods: Reports not retrieved | 2 |
| Other methods: Reports assessed for eligibility | 10 |
| Other methods: — Wrong population | 4 |
| Other methods: — No outcome data | 3 |
Title and figure style
Takes the title out of the image and sets the figure number in bold and the title in italics above it — the APA 7 layout. Credit the template in the note.
How to Use the PRISMA Flow Diagram Generator
- 1. Enter what you identified. Give each database and register on its own line with its count —
Embase, 1054— or just the totals. The template asks, where feasible, for the count from each source rather than one total, and the diagram lists the sources you give. - 2. Enter what was removed and excluded. Duplicates, records marked ineligible by automation tools, records removed for other reasons, records excluded at title and abstract, reports not retrieved, and the reports excluded at full text — one reason per line with its count.
- 3. Leave the in-between steps blank. Records screened, reports sought, reports assessed and reports of included studies are worked out and shown as placeholders. Type a number only where your own records say something else, and the tool tells you if the two disagree.
- 4. Add the other-methods column if you used it. Websites, organisations, citation searching and any other source go in the right-hand column, which joins the main flow at Studies included in review.
- 5. Download. Choose template colors or black and white, keep or drop the template's footnotes and source line, and download SVG or PNG. Copy counts puts every box and its number on the clipboard as a table.
Records, Reports and Studies
PRISMA 2020 counts three different things, and the boxes change from one to the next. A record is the title or abstract of a report as it appears in a database or on a website. A report is a document about a study: a journal article, preprint, conference abstract, study register entry, clinical study report, dissertation, unpublished manuscript or government report. A study is the investigation itself — and one study can have several reports, which is why the final box gives both numbers. Screening works on records; retrieval and eligibility work on reports; the review includes studies.
| Box | What it counts |
|---|---|
| Records identified | Every record found, by database and register |
| Records removed before screening | Duplicates, records marked ineligible by automation tools, and other removals |
| Records screened | Records whose titles and abstracts were read |
| Records excluded | Records ruled out at title and abstract — no reasons needed |
| Reports sought for retrieval | Full texts you tried to obtain |
| Reports not retrieved | Full texts you could not obtain |
| Reports assessed for eligibility | Full texts read against the inclusion criteria |
| Reports excluded | Full texts ruled out, with each reason and its count |
| Studies included in review | Studies in the review, and the reports of them |
The Arithmetic That Has to Add Up
Each step is the one before it minus what left the flow. Records screened = records identified − records removed before screening; reports sought = records screened − records excluded; reports assessed = reports sought − reports not retrieved; and reports of included studies = reports assessed − reports excluded, summed over both columns. Mismatches usually mean a count was taken at a different moment, or a duplicate was removed twice. The tool also checks that the exclusion reasons do not add up to more than the reports assessed, and that the number of studies is not larger than the number of reports.
The example review, worked through
- Records identified: 812 + 1,054 + 391 = 2,257 from databases and 31 + 17 = 48 from registers, 2,305 in all
- Records screened: 2,305 − 612 duplicates − 87 marked by automation tools − 0 other = 1,606
- Reports sought: 1,606 − 1,498 excluded at title and abstract = 108
- Reports assessed: 108 − 6 not retrieved = 102
- Reports excluded: 31 + 22 + 17 + 11 = 81, leaving 102 − 81 = 21 reports
- Other methods: 12 sought − 2 not retrieved = 10 assessed; 10 − 7 excluded = 3 reports
- Reports of included studies: 21 + 3 = 24, describing 19 studies
Which Template, and What to Leave Out
The PRISMA website offers four flow diagram templates: for new reviews and for updated reviews, each with or without the other-methods column. This tool draws the two for new reviews. An updated review adds a column for the studies in its previous version, so use the official template for that. A box or a line that does not apply to your review can be removed — the template says so — and Leave out lines you left blank does it for you. If automation tools excluded records, the template's second footnote asks you to say how many were excluded by a person and how many by the tool.
The flow diagram answers item 16a of the PRISMA 2020 checklist: describe the results of the search and selection process, from the records identified to the studies included, ideally with a flow diagram. The template is published under a CC BY 4.0 license, which allows it to be adapted as long as the original is credited. The tool keeps the template's source line under the diagram; in an APA paper, also credit Page et al. (2021) in the figure note and list the statement in your references. The statement is open access at doi.org/10.1136/bmj.n71, and the official templates are at prisma-statement.org.
Frequently Asked Questions
- What is the difference between records, reports and studies?
- A record is a title or abstract as indexed in a database or website, a report is a document describing a study, and a study is the investigation itself. Duplicates are records of the same report. One study can have several reports — a trial with a main paper and a follow-up — so the final box gives the number of studies and the number of reports.
- Where do duplicates go in a PRISMA flow diagram?
- In the box beside Records identified, titled Records removed before screening, on the line Duplicate records removed. They come off before screening, so records screened is the total identified minus the duplicates and the other records removed at that stage.
- Do I need reasons for every exclusion?
- For reports excluded at full text, yes: the template lists each reason with its count, and the reasons should add up to the total excluded. Records excluded at the title-and-abstract stage need only a number. List the reasons in the order your protocol uses, or by count.
- What if I didn't search registers or use automation tools?
- Leave those boxes blank and keep Leave out lines you left blank ticked: the line disappears from the diagram, which the template allows for steps that do not apply. A blank removal counts as zero in the arithmetic. If a step did happen but removed nothing, type 0 so the diagram says so.
- Can I use this for a scoping review or an updated review?
- Scoping reviews usually report against PRISMA-ScR, which also asks for the numbers screened, assessed and included with reasons for exclusion, and a flow diagram in this layout often serves; check what your journal expects. An updated systematic review needs the PRISMA 2020 template with the previous-version column, available on the PRISMA website.
- Is the PRISMA 2020 template free to use?
- Yes. The flow diagram templates are distributed under a Creative Commons Attribution 4.0 license, which lets anyone use and adapt them, provided the PRISMA 2020 statement (Page et al., 2021, BMJ 372:n71) is credited. Keep the source line, or credit the statement in your figure note.