The PRISMA 2020 Flow Diagram: Content of Each Stage

The PRISMA 2020 flow diagram is the standard figure for reporting how the pool of records in a systematic review was narrowed to the studies included. It is available in two templates: one for reviews that searched only databases and registers, and one for reviews that also searched other sources, such as citation lists or websites.

The stages in sequence

The PRISMA 2020 flow diagram documents identification, screening and inclusion, per the PRISMA 2020 statement1:

1. Identification — records retrieved from databases and registers (plus, in the other-methods template, records from other sources), with duplicates removed and records excluded by automation tools recorded separately.
2. Screening — titles and abstracts screened, with the number excluded; then reports sought for retrieval, with the number not retrieved recorded separately; then reports assessed for eligibility against the full text.
3. Screening: full-text exclusions — reports excluded at full-text assessment, each exclusion reason listed with its own count (rather than combined into a single “excluded” number).
4. Included — studies included in the review, and separately, reports of included studies (since one study can be reported across multiple publications).

Reconciliation of the reported counts

Every count in the diagram should trace to the count before it: records identified, minus duplicates and automation exclusions, equals records screened; records screened, minus those excluded at title/abstract, equals reports sought for retrieval; reports sought, minus those not retrieved, equals reports assessed for eligibility; reports assessed, minus those excluded (with reasons), equals studies included. A diagram in which these numbers do not reconcile is among the quickest reasons for a methods reviewer to flag a systematic review for revision.

The ‘other methods’ template

PRISMA 2020 provides a second template for reviews that supplement database searching with other methods — citation searching, hand-searching, website or organisation searching, or contacting study authors. In that template, “other methods” records run through their own identification and screening counts in parallel with the database/register stream, and both streams feed into the same included stage1.

Key points

Practical application

Frequently asked questions

@misc{prisma_study_selection_flow_2026,
  author       = {{Pubrica Academy}},
  title        = {The PRISMA 2020 Flow Diagram: Content of Each Stage},
  howpublished = {Pubrica},
  year         = {2026},
  month        = {oct},
  url          = {https://pubrica.com/academy/flowcharts/prisma-study-selection-flow/},
  note         = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}

Records identified (databases/registers): ___
Records identified (other methods): ___
Duplicates removed: ___
Excluded by automation tools: ___
Records screened: ___
Records excluded at title/abstract: ___
Reports sought for retrieval: ___
Reports not retrieved: ___
Reports assessed for eligibility: ___
Reports excluded, by reason: ___
Studies included: ___
Reports of included studies: ___