A PRISMA flow diagram is the four-phase chart that traces every record through a systematic review, from the total identified by the search, through the numbers screened and excluded, to the studies included. This generator takes the counts you captured at each stage, derives the figures that must follow from them, checks that the totals reconcile, and renders a clean PRISMA 2020 diagram you can download as a scalable SVG.
PRISMA flow diagram generator
Enter the record counts from each stage of study selection and download a reconciled PRISMA 2020 flow diagram ready to drop into your manuscript.
1215 records identified, 43 studies included. The PNG renders at three times figure size for print.
Report-ready text
Database and register searches identified 1,215 records; after removal of 320 duplicate records, 895 records were screened and 760 excluded. Of 135 reports sought, 12 could not be retrieved; 123 were assessed for eligibility and 80 were excluded, leaving 43 studies in the review.
The figure a reviewer checks first
The PRISMA flow diagram is usually the first thing an editor or peer reviewer turns to, because it shows at a glance whether your study selection was complete and honest. Every record your search returned has to be accounted for: either it was removed as a duplicate, excluded at title and abstract screening, could not be retrieved, was excluded at full text with a stated reason, or made it into the review. There is no room for records to disappear silently between phases. That is why the diagram is best understood as an arithmetic statement rather than a decoration, and why this tool derives the records screened, reports sought, reports assessed, and studies included rather than letting you type numbers that might not add up. If you want the underlying logic explained in full, our guide to the anatomy of the PRISMA flow diagram walks through each box.
What to enter, and what is calculated
You enter only the counts that come from a decision: how many records each database and register returned, how many duplicate records you removed, how many were set aside by automation tools or for other reasons, how many you excluded at screening, how many reports you could not retrieve, and how many you excluded at full text. The tool then computes the running totals so the four phases reconcile. Capturing these numbers as you go, rather than reconstructing them at the end, is the single best way to avoid a diagram that a reviewer sends back, a point we make in our walkthrough of the stages of a systematic review and in the practical advice on removing duplicate records cleanly.
A worked example and the mistakes to avoid
Suppose your databases return 1,200 records and a trials register adds 15, for 1,215 identified. You remove 320 duplicates, leaving 895 screened. Title and abstract screening excludes 760, so you seek 135 reports; 12 cannot be retrieved, leaving 123 assessed at full text. You exclude 80 with reasons, and 43 studies remain in the review. Every number on the right of the diagram is mirrored by a step down the left, and the tool will warn you the moment a count cannot be reconciled, for instance if you claim to have excluded more reports than you assessed. The most common errors are quietly dropping records between identification and screening, recording exclusion reasons at the screening phase rather than at full text, and mixing the 2020 layout with the older 2009 one. When the excluded full texts need to be tied back to your protocol, it helps to have pre-specified the inclusion and exclusion criteria that each reason maps to, and to have agreed how many reviewers screen each record before you begin.
Where the diagram fits in the report
The flow diagram is a required item of the PRISMA 2020 reporting checklist, and it sits alongside the narrative description of your search and selection. Once your included set is fixed, the same counts feed straight into the rest of the review, from extracting data from the included studies to the synthesis itself. A diagram that reconciles is a small but unmistakable signal that the rest of the methods were run with the same care.
How it works
The tool treats the diagram as a chain of subtractions. The records identified are the sum of the database and register counts, and the records removed before screening are the sum of duplicates, automation exclusions, and other removals.
identified = databases + registers
screened = identified − (duplicates + automation + other)
sought = screened − records excluded
assessed = sought − reports not retrieved
included = assessed − reports excluded at full text
Each derived value must be zero or greater. If any subtraction would go negative, the diagram is not drawn and the offending count is flagged, so the figure you download is always internally consistent. The layout follows PRISMA 2020 for studies identified through databases and registers.
Frequently asked questions
- What are the four phases of the PRISMA flow diagram?
- Identification, where records are found through databases and registers and duplicates are removed; screening, where titles and abstracts are assessed and the bulk of records are excluded; eligibility, where full-text reports are sought, retrieved, and assessed; and inclusion, the final count of studies that enter the review. This generator lays out all four phases and fills in the counts you cannot, or should not, type by hand.
- Why do the numbers in a PRISMA flow diagram have to add up?
- The diagram is an arithmetic statement: the records entering each phase, minus those excluded, must equal the number passing to the next. A figure that does not reconcile makes a reviewer doubt the whole review. The generator enforces this for you by deriving the screened, sought, assessed, and included counts from the numbers you enter, and it refuses to draw a diagram whose totals would go negative.
- How does the PRISMA 2020 flow diagram differ from the 2009 version?
- The 2020 version separates records removed before screening into duplicates, automation exclusions, and other removals, and it distinguishes records from reports throughout the lower phases. This tool follows the 2020 layout for studies identified through databases and registers, which is the version journals now expect for a new systematic review.
- At which stage do you record reasons for exclusion?
- Only at the eligibility, or full-text, phase. Screening exclusions are reported as a single count, while each excluded full-text report is attributed to a specific reason category. The optional reasons box in this tool lists those categories beneath the diagram, exactly where PRISMA expects them.
More review and meta-analysis tools
Meta-analysisPool study effect estimates with fixed-effect and random-effects models, with the heterogeneity statistics shown.Open tool Heterogeneity (I-squared)Quantify how much your studies disagree: Cochran's Q, I-squared, H, and the between-study variance tau-squared.Open tool Odds ratio / risk ratioTurn a 2x2 table of events into an odds ratio, risk ratio, and risk difference, each with a 95% confidence interval.Open tool