PRISMA 2020 Flow Diagram Generator

1. **Identification**: enter the number of records found in **databases** and in **registers** (such as trial registries). PRISMA 2020 recommends also reporting the count for each database or register separately, which you can list in the box or in your methods text. 2. **Records removed before screening**: enter the number of **duplicate records removed**. 3. **Screening**: enter the number of **records screened** (titles and abstracts) and the number of **records excluded** at that stage. 4. **Retrieval**: enter the number of **reports sought for retrieval** and the number **not retrieved** (for example, full text unavailable). 5. **Eligibility**: enter the number of **reports assessed for eligibility** and the **reports excluded**, broken down **with reasons** (each reason and its count). 6. **Included**: enter the number of **studies included in the review**. 7. Check the preview, then download the diagram as **SVG** (sharp at any size, best for journal submission and further editing) or **PNG** (easy to paste into Word or slides).

PRISMA stands for **Preferred Reporting Items for Systematic reviews and Meta-Analyses**. It is a reporting guideline: it tells authors what to report so that readers can see how a systematic review was done and judge how trustworthy it is. It does not tell you how to conduct the review.

The current version is the **PRISMA 2020 statement** (Page et al., *BMJ*, 2021), which replaced the PRISMA 2009 statement. It includes a 27-item checklist, a checklist for abstracts, and flow diagram templates. Many journals that publish systematic reviews ask authors to follow PRISMA and to submit the completed checklist and flow diagram.

The flow diagram shows how many records were found, how many were removed at each step and why, and how many studies ended up in the review. Separate official templates exist for new reviews that searched only databases and registers, for new reviews that also used other sources (such as websites, organisations, or citation searching), and for updated reviews. The original templates are available from the PRISMA website (prisma-statement.org).

PRISMA 2020 uses three terms precisely, and mixing them up is the most common source of confusing diagrams:

• A **record** is the title and/or abstract of a report as indexed in a database or website. Records are what you deduplicate and screen.
• A **report** is a document that supplies information about a study, such as a journal article, preprint, conference abstract, thesis, or trial register entry. Reports are what you retrieve and assess in full text.
• A **study** is an investigation, such as a clinical trial, which may be described in several reports.

This is why the upper boxes count records, the middle boxes count reports, and the final box counts studies. If several reports describe the same study, the number of included studies can be smaller than the number of included reports; the official template has a separate line for "reports of included studies".

| Stage | Box | What to count | |---|---|---| | Identification | Records identified from databases / registers | All records returned by your searches, before deduplication | | Identification | Records removed before screening | Duplicates removed (the official template also lists records marked ineligible by automation tools and records removed for other reasons) | | Screening | Records screened | Records that went through title/abstract screening | | Screening | Records excluded | Records excluded at title/abstract stage | | Screening | Reports sought for retrieval | Full texts you tried to obtain | | Screening | Reports not retrieved | Full texts you could not obtain | | Screening | Reports assessed for eligibility | Full texts read against your eligibility criteria | | Screening | Reports excluded, with reasons | Each reason with its count | | Included | Studies included in review | Final number of studies |

The numbers should add up from one box to the next. A worked set of counts:

• Databases 1,248 + registers 52 = **1,300** records identified
• 312 duplicates removed → **988** records screened
• 871 records excluded → **117** reports sought for retrieval
• 6 not retrieved → **111** reports assessed for eligibility
• 87 excluded (wrong population 34, wrong outcome 29, wrong study design 24) → **24** studies included

Here each included report is assumed to describe a different study. If some studies have more than one report, explain the difference between reports and studies in the text.

• **Numbers that do not add up.** Every exclusion should account for the drop between boxes. Reviewers check this.
• **Excluding at full-text stage without reasons.** PRISMA 2020 expects reasons for excluded reports, with counts.
• **Mixing records, reports, and studies**, for example reporting "studies screened" when you screened records.
• **Counting after deduplication as "identified".** The identification box should show what the searches returned.
• **Using the 2009 template** with labels such as "full-text articles assessed" when the journal asks for PRISMA 2020.
• **Leaving out registers or other sources** that you actually searched. If you used citation searching or websites, the official template with the "other methods" column shows these separately; describe them in your methods even if your diagram uses the simpler layout.
• **Low-resolution images.** Export as SVG, or as a high-resolution PNG, so text stays legible in print.

In the methods section, state that the review is reported following PRISMA 2020, and in the results section refer to the flow diagram, for example: "The study selection process is shown in Figure 1 (PRISMA 2020 flow diagram)." Most journals expect the completed PRISMA checklist as a supplementary file.

Cite the statement itself:

> Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. *BMJ*. 2021;372:n71. doi:10.1136/bmj.n71

Format the reference in your journal's required style.

If you are still at the stage of mapping the literature, Data2Paper's [Research Report](/research-report) tool can draft a structured literature review on your topic using real papers from OpenAlex and Crossref with checkable DOIs, and you can preview the first part for free. It is a starting point for scoping, not a substitute for a systematic search and screening process documented with PRISMA.