A PRISMA flow diagram shows how many records your systematic review identified, screened, excluded and included, in the PRISMA 2020 layout. Paste your screening counts and Vizcept lays out the boxes and exclusion reasons; check each number against your log, because an image model draws the figure.
Attach a sketch or an existing image (JPG/PNG/WEBP, up to 10MB)
Free trial: 5 credits, 2K resolution with watermark
Enter your screening numbers on the left to generate your own PRISMA flow diagram
You already have the numbers in your screening spreadsheet. The slow part is redrawing the boxes and arrows every time a reviewer sends the counts back. Paste the numbers as a sentence and let the generator lay out the three phases.
Type the numbers in plain English: records identified, duplicates removed, records screened and excluded, reports sought and not retrieved, full texts assessed, exclusions with reasons, and studies included. Give each database its own count if you want them listed separately.
The generator arranges Identification, Screening, and Included as stacked bands, routes the main flow downward, and sends every exclusion to a box on the right β the structure the PRISMA 2020 statement specifies.
Read the rendered counts against your spreadsheet before you use the figure. Download PNG at 300 DPI for Word and journal submission, or SVG when you need to scale it for a poster.
A PRISMA flow diagram is the figure in a systematic review that reports how many records were identified, screened and excluded, and how many studies were included, and why full-text reports were excluded. PRISMA stands for Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Item 16a of the PRISMA 2020 checklist asks authors to describe the results of the search and selection process, "ideally using a flow diagram". The diagram counts three different things:
| Term | What it means in PRISMA 2020 |
|---|---|
| Record | The title or abstract of a report, as indexed in a database or website. The same report found twice is a duplicate. |
| Report | A document about a study: a journal article, preprint, conference abstract, register entry or dissertation. |
| Study | The investigation itself. One study can have several reports. |
PRISMA 2020 has four flow diagram templates. Pick by two questions: is this a new review or an update, and did you search only databases and registers, or other sources too?
| Your review | Template |
|---|---|
| New review, databases and registers only | One column: Identification β Screening β Included |
| New review, plus websites, organisations or citation searching | Two columns: databases and registers on the left, other methods on the right |
| Updated review, databases and registers only | Adds a box for studies included in the previous version |
| Updated review, plus other sources | Two columns plus the previous-version box |
Mention your sources and whether it is an update in the description, and the layout follows. Scoping reviews usually adapt the same flow, reporting "sources of evidence". Cite the diagram's source as: Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.
Reviewers check that every box follows from the one above it. Run these subtractions on your screening log before you generate, and again on the finished figure β an image model can change a digit.
Three rules keep the boxes consistent:
A worked example:
| Box | n | Check |
|---|---|---|
| Records identified (1,190 databases + 58 registers) | 1,248 | |
| Removed before screening (duplicates) | 312 | 1,248 β 312 = 936 |
| Records screened | 936 | |
| Records excluded | 812 | 936 β 812 = 124 |
| Reports sought for retrieval | 124 | |
| Reports not retrieved | 6 |
Elsevier, Wiley and Taylor & Francis list flowcharts or flow diagrams among the figures AI may help with, with disclosure, and several want the disclosure in the caption. Springer Nature does not permit content created from text prompts, so ask the editor there. See the .
A PRISMA diagram also carries your screening counts, so check every number against your screening log before you submit. Writing a thesis? The covers the framework chapter too.
The five common ways to make the figure differ in what you type and who places the numbers.
| Option | Cost | What you enter | Download | Who places the numbers |
|---|---|---|---|---|
| PRISMA 2020 Word templates | Free (CC BY 4.0) | Every n, typed into the boxes | Word (.docx), four versions | You |
| PRISMA2020 Shiny app and R package | Free, open source | A data-entry form or CSV | HTML (interactive), PDF, PNG, SVG and more | The app, from your form |
| Covidence | $339 a year for one review, $907 for up to three, or an institutional licence; free trial with 500 records | Your sources at import, or each database's count if you imported from a citation manager | Word (.docx), or copy the counts as text | Covidence, from your screening |
| Rayyan | Essential, Advanced and institutional plans | Databases, registers, not-retrieved, assessed and included counts; Rayyan fills duplicates and screening counts | PDF, PNG, SVG, interactive HTML | Rayyan and you |
| Vizcept | Free credits on sign-up; a one-time pack of 150 credits is $19.90, valid 365 days, no renewal |
Students Β· Researchers Β· Methods teachers
A PRISMA flow diagram is a reporting requirement, not a design choice. These are the three groups that hit that requirement most often.
First systematic review, hard deadline
Your supervisor asked for a PRISMA diagram and you have never drawn one. You know your counts; you do not know which box the 'reports not retrieved' number belongs in. Start from a correct layout instead of a blank canvas.
Submission-ready figures, fast revisions
Reviewer 2 asked you to split the database column from citation searching, and your counts shifted after the update search. Regenerate the whole figure from a revised sentence rather than nudging boxes in PowerPoint.
Teaching evidence synthesis
You teach search strategy and screening to cohorts who have never seen a flow diagram. Produce a blank template for a handout, then a filled example with realistic counts, without opening drawing software.
Six review designs
Each figure below was generated from the text shown underneath it. Reuse an example as a starting point and swap in your own counts.
Databases and registers only, single column, three phases
Two-column PRISMA 2020 with citation searching on the right
PRISMA-ScR style with grey literature and charted sources
PRISMA 2020 renamed the first box and dropped the second; use this layout only when a journal asks for it
Per-database record counts listed inside the identification box
Empty count boxes for a class handout or thesis appendix
The PRISMA 2020 flow diagram has a fixed anatomy: three phases, a downward main path, and exclusions branching right. This generator starts from that anatomy instead of a blank drawing canvas.
Records, reports and studies
Describe your review in English and the instructions ask for the official PRISMA 2020 box wording β 'Records screened', 'Reports sought for retrieval', 'Studies included in review'. Write in Spanish or Portuguese and the labels follow your language. Proofread them along with the numbers.
Every account gets free credits on sign-up, enough to draft your first PRISMA flow diagram. Upgrade for 4K, no watermark and more generations while you iterate through screening rounds β or buy a one-time credit pack for a single review.
Student or teacher? Apply for an education discount
Feedback from people who had to put a PRISMA flow diagram into a thesis, a protocol, or a submission.
Dr. Amara Okonkwo
Research Fellow, Public HealthOur update search added 400 records three weeks before submission. Regenerating the flow diagram from a revised sentence took under a minute; redrawing it by hand would have taken an afternoon.
Priya Raghunathan
PhD Candidate, NursingI could never remember whether 'reports not retrieved' sat before or after the eligibility box. Starting from a correct PRISMA 2020 layout meant I was checking numbers instead of second-guessing the structure.
TomΓ‘s Herrera
Medical LibrarianFAQ
Turn your screening counts into a PRISMA 2020 figure, check the numbers with the counts checklist, and export PNG at 300 DPI or SVG.
| 124 β 6 = 118 |
| Reports assessed for eligibility | 118 |
| Reports excluded: wrong population 41, wrong outcome 33, not an RCT 23 | 97 | 41 + 33 + 23 = 97; 118 β 97 = 21 |
| Reports of included studies | 21 |
If one study has several reports, the number of included studies can be smaller than the number of reports β say both. .
| One sentence with your counts |
| PNG at 300 DPI (watermarked on free credits); SVG on paid plans and packs |
| An image model; check every number |
The Word templates and the Shiny app place each number exactly where you type it. With Vizcept, read every number on every version. Prices checked 19 September 2026.
One line per database
Name each database with its count β MEDLINE n=812, Embase n=644, Scopus n=421 β and each gets its own line in the identification box. The PRISMA 2020 template asks for per-database counts where feasible.
Itemised, not lumped together
Write 'excluded with reasons: wrong population n=41, wrong outcome n=33, wrong study design n=18' and each reason appears as its own line inside the exclusion box. Count each report once, under its primary reason, so the lines add up to the total.
Update search, new figure
An update search or a second screener changes half your numbers. Edit the sentence and generate again rather than dragging boxes and re-typing every n value in a drawing tool.
Word, LaTeX, posters
PNG at 300 DPI drops straight into a Word manuscript or slide deck at print quality. SVG scales to conference-poster size without softening the text, for journals that ask for vector figures.
Scoping, meta-analysis, blank template
Scoping reviews chart sources instead of including studies; updated reviews carry studies over from the previous version; a journal that still wants the 2009-style meta-analysis boxes can have them; teaching handouts need empty boxes. Say which one you need and the labels change to match.
I support about thirty review teams a year. Being able to hand someone a blank template and a worked example in the same session has made the search-strategy consultations much shorter.
Dr. Amara Okonkwo
Research Fellow, Public HealthOur update search added 400 records three weeks before submission. Regenerating the flow diagram from a revised sentence took under a minute; redrawing it by hand would have taken an afternoon.
Priya Raghunathan
PhD Candidate, NursingI could never remember whether 'reports not retrieved' sat before or after the eligibility box. Starting from a correct PRISMA 2020 layout meant I was checking numbers instead of second-guessing the structure.
TomΓ‘s Herrera
Medical LibrarianI support about thirty review teams a year. Being able to hand someone a blank template and a worked example in the same session has made the search-strategy consultations much shorter.
Dr. Amara Okonkwo
Research Fellow, Public HealthOur update search added 400 records three weeks before submission. Regenerating the flow diagram from a revised sentence took under a minute; redrawing it by hand would have taken an afternoon.
Priya Raghunathan
PhD Candidate, NursingI could never remember whether 'reports not retrieved' sat before or after the eligibility box. Starting from a correct PRISMA 2020 layout meant I was checking numbers instead of second-guessing the structure.
TomΓ‘s Herrera
Medical LibrarianI support about thirty review teams a year. Being able to hand someone a blank template and a worked example in the same session has made the search-strategy consultations much shorter.
Dr. Amara Okonkwo
Research Fellow, Public HealthOur update search added 400 records three weeks before submission. Regenerating the flow diagram from a revised sentence took under a minute; redrawing it by hand would have taken an afternoon.
Priya Raghunathan
PhD Candidate, NursingI could never remember whether 'reports not retrieved' sat before or after the eligibility box. Starting from a correct PRISMA 2020 layout meant I was checking numbers instead of second-guessing the structure.
TomΓ‘s Herrera
Medical LibrarianI support about thirty review teams a year. Being able to hand someone a blank template and a worked example in the same session has made the search-strategy consultations much shorter.
Dr. Lena Fischer
Senior Lecturer, Evidence SynthesisI generate a blank version for the worksheet and a filled version for the answer key. My students stop fighting the drawing software and actually think about what each count means.
Michael Adeyemi
MSc Student, EpidemiologyMy supervisor asked for the two-column version with citation searching split out. I described the second column in one sentence and got the layout I had spent two evenings trying to build manually.
Dr. Sofia Almeida
Postdoc, Clinical PsychologyThe exclusion box with itemised reasons is the part reviewers always query. Having each reason on its own line with its own count saved a round of revisions.
Dr. Lena Fischer
Senior Lecturer, Evidence SynthesisI generate a blank version for the worksheet and a filled version for the answer key. My students stop fighting the drawing software and actually think about what each count means.
Michael Adeyemi
MSc Student, EpidemiologyMy supervisor asked for the two-column version with citation searching split out. I described the second column in one sentence and got the layout I had spent two evenings trying to build manually.
Dr. Sofia Almeida
Postdoc, Clinical PsychologyThe exclusion box with itemised reasons is the part reviewers always query. Having each reason on its own line with its own count saved a round of revisions.
Rahul Menon
Systematic Review CoordinatorWe run several reviews in parallel and the figures used to look different depending on who drew them. Now every protocol in the portfolio uses the same structure.
Dr. Hannah Whitfield
Assistant Professor, Health Services ResearchFor a scoping review the labels have to say charted sources rather than included studies. Asking for the PRISMA-ScR variant got the right wording without me editing text boxes.
Kwame Boateng
DrPH Candidate, Global HealthI export SVG for the poster and PNG for the thesis chapter. Same figure, both formats, no quality loss when it gets blown up to A0.
Explore more AI diagram tools across subjects β from the same discipline and beyond.
One subscription unlocks every Vizcept AI diagram tool β from free body diagrams and syntax trees to biology diagrams and more.
