PRISMA 2020 Checklist and Guidelines: All 27 Items
The 27 PRISMA 2020 items for systematic reviews and meta-analyses, what reviewers object to in each block, and a checker for your manuscript.
PRISMA 2020 is the guideline for reporting a systematic review, with or without meta-analysis, so that readers can see what was searched, what was excluded and why. It has 27 items in seven sections, a flow diagram and a separate 12-item checklist for the abstract. It replaces PRISMA 2009.
By Víctor Ciudad-Fernández, PhD · Updated: October 6, 2026 · Source: Page et al. (2021)
What PRISMA 2020 is and when to use it
PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) brings together what the article of a systematic review has to contain so that the search can be reproduced, the selection understood and the synthesis trusted. The 2020 version, by Page and colleagues, was published in 2021 with an explanation and elaboration paper that gives examples for each item.
People sometimes call it "the PRISMA method", but it is not a method for doing the review: it is a guideline for reporting it. How to search, screen, assess risk of bias and synthesize is set by the method you follow (for example, the Cochrane Handbook) and your protocol; PRISMA tells you what to report about each step.
Systematic reviews of interventions, with or without meta-analysis: the case it is written for.
Other kinds of systematic review (etiology, prognosis, prevalence): almost all of it applies, and there are extensions for specific cases.
Scoping reviews: PRISMA-ScR. The review protocol: PRISMA-P. Documenting the search in detail: PRISMA-S.
A narrative review without a systematic search is not a PRISMA review, even if you cite the guideline.
The 27 items, section by section
Several items have sub-items (10a and 10b, 13a to 13f, 20a to 20d…). What follows explains each one in plain words; for submission, use the official checklist. Under each section you have what reviewers usually object to in it.
Title and abstract
1. The title identifies the report as a systematic review.
2. An abstract that follows the PRISMA 2020 for Abstracts checklist.
What reviewers usually object to:
The title says "review" without "systematic", or "meta-analysis" when no results were pooled.
The abstract does not give the number of included studies and participants or the certainty of the evidence.
Introduction
3. The rationale for the review in the context of what is already known.
4. The question or objectives, stated explicitly.
What reviewers usually object to:
No previous review on the same question is cited, or what this one adds is not explained.
The question is not framed as PICO (or an equivalent) and the eligibility criteria do not match it.
Methods
5. The inclusion and exclusion criteria, and how studies were grouped for the syntheses. (checked by the checker)
6. Every source searched (databases, registers, websites, organizations, reference lists…) with the date each was last searched. (checked by the checker)
7. The full search strategy for every database, register and website, with filters and limits. (checked by the checker)
8. How it was decided whether a study met the criteria: how many people screened each record and report, whether they worked independently, and whether any automation tool was used. (checked by the checker)
9. How data were collected: how many people, whether independently, how information was requested or confirmed with authors, and whether automation was used.
10a. Every outcome sought and, if not all results compatible with each one (measures, time points, analyses) were collected, how they were chosen.
10b. The other variables collected (participants, intervention, funding…) and the assumptions made about missing or unclear information.
11. How the risk of bias of the studies was assessed: with which tool, by how many people, and whether independently. (checked by the checker)
12. The effect measure for each outcome (mean difference, risk ratio…).
13a. How it was decided which studies went into each synthesis.
13b. How data were prepared for synthesis (missing statistics, conversions).
13c. How the results of studies and syntheses were tabulated or displayed.
13d. The synthesis methods and why they were chosen; for a meta-analysis, the model, how heterogeneity was handled and the software used.
13e. How possible causes of heterogeneity were explored (subgroups, meta-regression).
13f. Any sensitivity analyses.
14. How the risk of bias due to missing results (publication or reporting bias) was assessed.
15. How the certainty (or confidence) in the body of evidence for each outcome was assessed.
What reviewers usually object to:
The search strategy is a list of keywords, not the full syntax for each database: it cannot be reproduced.
The last search is more than a year old when the manuscript is submitted.
There is no statement that two people screened independently, or how disagreements were resolved.
Risk of bias is assessed with a quality scale that adds up points instead of a domain-based tool such as RoB 2 or ROBINS-I.
The fixed- or random-effects model is chosen depending on how heterogeneity turns out, without prior justification.
Results
16a. The results of the search and selection, from records identified to studies included, ideally with a flow diagram. (checked by the checker)
16b. Studies that seemed to meet the criteria but were excluded, with the reason.
17. Each included study, cited and with its characteristics.
18. The risk-of-bias assessment for each included study.
19. For each outcome, the summary statistics for each group and the effect estimate with its precision, ideally in tables or a forest plot.
20a. For each synthesis, the characteristics and risk of bias of the contributing studies.
20b. The results of each synthesis; for a meta-analysis, the pooled estimate with its confidence interval and the measures of heterogeneity.
20c. The results of the investigation of heterogeneity.
20d. The results of the sensitivity analyses.
21. The risk of bias due to missing results for each synthesis.
22. The certainty of the evidence for each outcome.
What reviewers usually object to:
The flow diagram does not add up: excluded plus included does not equal screened.
There is no list of studies excluded at full text, with reasons.
I² is reported without a prediction interval and with no attempt to explain heterogeneity.
Publication bias is assessed with a funnel plot of fewer than ten studies.
Discussion
23a. A general interpretation of the results in the context of other evidence.
23b. The limitations of the evidence included.
23c. The limitations of the review process.
23d. The implications for practice, policy and future research.
What reviewers usually object to:
The conclusion ignores the low certainty of evidence that the review itself rated.
The limitations of the included studies are mixed up with those of the review.
Other information
24a. The registration of the review (registry name and number) or a statement that it was not registered. (checked by the checker)
24b. Where the protocol can be accessed, or a statement that none was prepared.
24c. Any changes from the registration or the protocol, with an explanation.
25. The sources of funding and the role of the funders.
26. The authors’ competing interests.
27. Which materials are publicly available and where: data extraction forms, extracted data, data used in the analyses, code and anything else.
What reviewers usually object to:
The review was not registered in PROSPERO or OSF before starting, or what is published does not match the registration and this is not explained.
The extracted data table is not shared, and it is what makes the meta-analysis repeatable.
Check your manuscript against PRISMA 2020
Paste the article (Methods and Results at a minimum). The checker searches the text for 6 of the 27 PRISMA 2020 items, the ones a text search can decide without reading (search strategy, databases searched, risk-of-bias assessment, study-selection flow diagram, protocol registration, eligibility criteria and selection by two reviewers), and shows you the sentence that supports each one. It runs in your browser: nothing is uploaded or stored.
Checklist checker: Detects the design and picks the guideline for you.
Previum · AI peer review. A checklist tells you whether something appears in the text; it does not tell you whether it was done right. Once the manuscript is complete, see what Reviewer 2 would object to before you submit: findings anchored to your text, in the margin. Review my manuscript
What changed from PRISMA 2009
PRISMA 2009 also had 27 items, but not the same ones. The 2020 version splits many into sub-items, adds how the risk of bias due to missing results and the certainty of the evidence were assessed, asks you to say which data and code are available, and asks for an explicit statement on whether the review was registered or not.
The flow diagram changed too: it now separates studies found through databases and registers from those found by other routes (reference lists, websites, contacting authors), and it has a version for updated reviews.
How to cite the guideline
If your Methods section says you followed PRISMA 2020, cite the statement (and the explanation and elaboration if you rely on it), not this page.
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., … Moher, D. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
Page, M. J., Moher, D., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw, J. M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., … McKenzie, J. E. (2021). PRISMA 2020 explanation and elaboration: Updated guidance and exemplars for reporting systematic reviews. BMJ, 372, n160. https://doi.org/10.1136/bmj.n160
No. It is a guideline for reporting one. The method (how to search, screen, assess risk of bias and synthesize) comes from your protocol, usually based on the Cochrane Handbook or the handbook of the relevant organization.
How many items does PRISMA 2020 have?
Twenty-seven, several with sub-items, plus a separate 12-item checklist for the abstract. PRISMA 2009 also had 27, but the numbering and content changed.
Does PRISMA work for a scoping review?
That is what PRISMA-ScR, its extension for scoping reviews, is for. It shares much of the structure, but critical appraisal of the sources is optional and effects are not pooled.
Where can I make the flow diagram?
You can use the PRISMA 2020 flow diagram generator on this site: enter the numbers for each stage, it computes the boxes that depend on others so they never contradict each other, and you download it as a PNG.