PRISMA REPORTING AND PUBLICATION
PRISMA 2020 reporting and systematic review publication
Turn a completed review into a transparent, publication-ready research report. Apply PRISMA 2020 across the manuscript, abstract, figures, tables, and supplementary files while preserving consistency between the protocol, analyses, results, and conclusions.
Reporting as methodological traceability
Reporting a systematic review as an auditable research record
A completed systematic review is not publication-ready simply because the searches, screening, appraisal and synthesis are finished. The final report has a different methodological task: it must allow another reader to reconstruct what the review set out to do, what was actually done, what evidence was found, how the analyses were produced, where important decisions changed, and how the stated conclusions follow from the results. PRISMA 2020 provides the central reporting framework for that task, but transparent publication requires more than placing a tick beside 27 checklist items.12
PRISMA 2020 remains the main PRISMA reporting guideline. It was developed primarily for systematic reviews evaluating effects of interventions and comprises the statement paper, a 27-item checklist, an expanded checklist, the PRISMA 2020 for Abstracts checklist and flow-diagram templates for original and updated reviews.1 The Explanation and Elaboration paper provides the rationale, detailed recommendations and examples needed to interpret those items appropriately.2 Neither document is a substitute for methodological guidance on how a review should be conducted.
Reporting quality is not methodological quality
A review may transparently report a weak method, incompletely report a sound method, or do both well. PRISMA concerns the completeness and transparency of the report. Conduct standards and methodological handbooks address how the review should be performed; critical-appraisal instruments address whether its methods create important risks to validity. These questions interact, but they are not interchangeable. Cochrane makes the distinction explicit: good conduct does not guarantee good reporting, good reporting cannot repair poor conduct, and poor reporting may obscure either.3
Why the final manuscript is not merely a writing exercise
A systematic review accumulates decisions over time. Eligibility criteria are defined, databases are searched, records are deduplicated and screened, information is extracted, study limitations are assessed, outcomes are organized, quantitative or qualitative syntheses are performed, and conclusions are developed. By the time authors begin the final manuscript, many of those decisions may be separated from one another across protocols, registration records, screening software, spreadsheets, statistical files, meeting notes and supplementary documents.
The publication problem is therefore one of reconstruction. The report has to restore the connections among these decisions without pretending that the review proceeded more neatly than it actually did. A reader should be able to move from the review question to the eligibility criteria, from eligibility to study selection, from the included studies to the analyses, and from those analyses to the claims made in the abstract and discussion.
This is why a checklist completed only after the manuscript is finished is a weak substitute for reporting control during writing. The PRISMA checklist is most useful when it functions as a map between reporting expectations and the places where evidence for those expectations can actually be found. That place may be the Methods section, a results table, the PRISMA flow diagram, a supplementary search strategy, a protocol, a registration record or a repository. What matters is that the information is reported completely enough to be found and interpreted.
Prespecification creates a reference point, not a prohibition on change
PRISMA 2020 asks authors to provide registration information, indicate where a protocol can be accessed and describe and explain amendments to information supplied in the registration or protocol.12 This requirement is important because the protocol records intended methods, while the completed report records what was actually done. The two documents serve different temporal functions.
A difference between them does not automatically indicate poor practice. Planned subgroup analyses may become impossible because too few studies are found. An outcome may require a different synthesis because the retrieved data have an unexpected structure. Eligibility wording may need clarification when an unforeseen class of study appears during screening. Cochrane explicitly recognizes that review authors may be unable to implement all planned methods and instructs authors to describe and explain amendments to prespecified methods.3
What threatens interpretability is an unexplained change, particularly when readers cannot determine when the change occurred or whether it may have been influenced by knowledge of the emerging results. Meta-research shows that protocol-to-publication differences are common. In a random sample of 100 non-Cochrane systematic reviews with PROSPERO protocols, 90% differed from their protocols in at least one eligibility component, yet the rationale for a component-level change was reported and explained in only one review.4 A later comparison of 97 Cochrane and 97 non-Cochrane reviews found changes in PICOS elements in 54.6% and 67.0% of reviews respectively; reporting of those changes was markedly more complete in the Cochrane sample.5
Those observations do not establish that every change was biased. They establish a reporting problem: without a visible audit trail, readers cannot distinguish a justified amendment from an undocumented deviation.
The systematic review report is a connected set of research objects
PRISMA 2020 does not require every detail to occupy the same physical page. Complete reporting often depends on several linked objects. Search strategies may be placed in supplementary files; extracted data and analytic code may be deposited in a repository; excluded studies may appear in an appendix; registration and protocol records may sit on external platforms; the flow diagram summarizes study identification and selection; and the abstract compresses the main report for readers who may never open the full article.
Distributed reporting is therefore not inherently incomplete reporting. The problem arises when the distribution becomes fragmentation: a search strategy cannot be found, a supplement is not clearly linked, a data-availability statement does not lead to usable material, or the abstract communicates a materially different interpretation from the full report.
This distinction matters for reproducibility. In a content analysis of 300 systematic reviews with meta-analysis, 29% included a data-availability statement and only 2% included both data- and code-availability statements.6 More recent reproduction research provides a complementary picture. In a sample of 296 systematic reviews, 121 meta-analyses required independent reproduction; 104 of those 121, or 86%, were fully reproducible under the study’s prespecified criteria, while ten could not be reproduced because usable data were unavailable.7 The finding should not be generalized into a claim that all meta-analyses are reproducible. It demonstrates why access to analysis-ready inputs and sufficiently specific reporting can determine whether independent checking is possible.
Replicability poses a different problem. Reproduction asks whether an analyst can obtain the same result from the same underlying inputs. Replication asks whether researchers repeating the described review process obtain comparable study sets and results. A 2026 REPRISE study found that independent repetition of literature searches and review processes could produce substantial discrepancies even when resulting meta-analytic conclusions often remained broadly similar.8 The distinction is useful because a manuscript may describe the statistical model well enough for calculation to be reproduced while describing study identification or eligibility decisions too weakly for the review process to be replicated.
Methods and Results have to form a one-to-one audit trail
A reporting checklist is easiest to misunderstand when it is treated as a sequence of isolated disclosures. The stronger approach is relational. Each important analytic result should have an identifiable methodological origin, and each major prespecified method should have a visible outcome in the report: it was performed and reported, not performed and explained, or amended and justified.
| Reporting relationship | What the reader should be able to establish | Typical failure |
|---|---|---|
| Eligibility criteria → included studies | Why each included study belongs to the review population, intervention/exposure, comparator, outcome and design scope. | Eligibility criteria change without explanation, or included studies appear inconsistent with stated criteria. |
| Search methods → identified records | Which sources were searched, when, how the strategies were constructed and how supplementary identification was performed. | Database names are supplied but the executed strategy cannot be reconstructed. |
| Synthesis methods → synthesis results | Which studies contributed to each synthesis, which effect measures and models were used and how heterogeneity or alternative analyses were handled. | A pooled result appears without enough information to understand the analysis that generated it. |
| Risk of bias/certainty → interpretation | How limitations in the evidence influence the strength and wording of the conclusions. | The Discussion makes stronger claims than the review’s own limitations support. |
| Full report → abstract | That a reader of the abstract reaches materially the same understanding as a reader of the full review. | Benefits are emphasized while important uncertainty, harms or limitations disappear. |
Cochrane’s current reporting chapter expressly recommends consistency of objectives, important outcomes, results, caveats and conclusions across the main text, tables and figures, abstract and other summaries.3 This is more than editorial tidiness. The most visible part of the review is often not the full Methods section but the title, abstract, forest plot, Summary of Findings table or conclusion. If those surfaces do not agree, the review can communicate different messages to different readers.
Misleading interpretation is one manifestation of this problem. A methodological classification identified 39 forms of spin in systematic reviews and meta-analyses, including misleading reporting, misleading interpretation and inappropriate extrapolation.9 The relevant reporting lesson is not to reduce all abstract problems to “spin.” Numerical inconsistency, omission and overinterpretation are different failures and should be audited separately.
PRISMA 2020 is the core, not the entire reporting ecosystem
The main PRISMA guideline was designed primarily around systematic reviews of intervention effects. Different review questions and methods can require additional reporting information. The PRISMA Executive therefore maintains extensions for particular review types and methodological components.14
For example, PRISMA-S provides reporting guidance for literature searches; its 16 items address aspects of searching that require more detail than the main checklist alone.10 PRISMA-LSR, published in 2024, operates as an add-on to PRISMA 2020 for living systematic reviews and many standard review updates.11 PRISMA-C 2026 is an extension of PRISMA 2020 for reviews of interventions involving children and adolescents.12 Other extensions cover areas including scoping reviews, network meta-analysis, individual participant data, diagnostic test accuracy, harms and outcome measurement instruments.
The practical lesson is not to memorize a static list. Reporting-guideline development continues. As of 8 August 2026, the official PRISMA register lists several extensions or updates under development, including rapid reviews, qualitative evidence synthesis, reviews using GRADE, prevalence reviews, health-economic evaluations, updates to PRISMA-NMA, PRISMA-ScR and PRISMA-P, and a partial update to incorporate guidance on the use of AI tools within the systematic-review process.14 Authors should therefore verify the current official PRISMA extension register for their review design rather than rely on an old checklist collection.
Current-methods note. Automation is not absent from PRISMA 2020: the expanded checklist already asks authors to describe automation tools used in the study-selection process where applicable.2 Because AI-supported review methods are evolving quickly and an official partial PRISMA update is under development, claims about additional AI-reporting requirements should be checked against the live PRISMA register before submission.14
Publication requirements add another layer
PRISMA does not replace the target journal’s instructions or broader editorial standards. Reporting guidance, journal policy and research governance overlap but have different authority. A journal may impose word limits, manuscript structure, data-sharing requirements, contributor statements, disclosure forms or policies on artificial-intelligence use that go beyond PRISMA.
The January 2026 ICMJE Recommendations, for example, address authorship, conflicts of interest, data access and use of artificial intelligence in publishing.13 These are editorial recommendations, not additional PRISMA checklist items. Keeping the distinction explicit prevents a common reporting error: attributing a publisher or editorial requirement to PRISMA simply because the two happen to ask for related information.
The publication-ready review therefore sits at the intersection of several controls: the review’s prespecified plan, the methods actually conducted, the appropriate reporting guideline and extensions, the journal’s instructions, and the integrity of the final research objects. The strongest report is the one in which those layers can be reconciled without forcing the reader to guess.
From reporting principles to working control
The resources and methodological guides below turn this reporting problem into concrete work. They are designed to help authors locate reporting gaps, construct a coherent manuscript and preserve the essential methods and findings when the review is compressed into an abstract.
This collection is intentionally extensible. Additional reporting resources can be added as review methods, journal expectations and PRISMA extensions evolve, while the underlying task remains the same: make the completed review traceable from its original plan to its published claims.
REPORTING AND PUBLICATION RESOURCES
Build and verify a transparent systematic review report
Move from a completed review to a transparent publication record. Audit PRISMA 2020 reporting, construct a coherent manuscript, and create a concise abstract that preserves the review’s methods, results, uncertainty, and conclusions.
Reporting resources for a completed systematic review
Use these practical resources to locate required PRISMA reporting, organize the full manuscript, and build an accurate abstract while maintaining consistency across the final research report.
PRISMA 2020 checklist and compliance tracker
Map every PRISMA 2020 item to the exact manuscript section, page, figure, table, or supplement where it is reported and identify unresolved reporting gaps.
Systematic review manuscript template
Draft and coordinate the title, introduction, methods, results, discussion, declarations, evidence displays, and supplementary materials as one coherent scientific report.
Systematic review abstract template and PRISMA checklist
Build a journal-ready abstract from verified full-report evidence, check the 12 PRISMA 2020 for Abstracts items, monitor the word limit, and verify numerical and interpretive fidelity.
PRISMA 2020 checklist explained: what to report and where
How to write a systematic review manuscript
How to write a systematic review abstract using PRISMA for abstracts
From reporting components to one coherent publication
Close the reporting loop before submission
The practical resources above solve different parts of the publication problem, but the final quality check has to return to the review as a whole. A checklist can locate missing information, a manuscript structure can organize that information, and an abstract framework can protect essential content under a severe word limit. None of them can substitute for the final act of reconciliation: checking that every public version of the review represents the same methods, evidence and level of certainty.
This final reconciliation is where reporting becomes more than documentation. It is an opportunity to identify discrepancies that accumulated during the review process. A result may have changed after data checking. A subgroup analysis may have been abandoned because too few studies were available. The final set of eligible studies may differ from the protocol’s anticipated scope. A figure may have been regenerated after an analytic correction while the accompanying paragraph retained an older number. A supplementary search strategy may no longer correspond to the database search date reported in the main manuscript. Each inconsistency is small in isolation; together they determine whether the report can be audited.
The final reporting task is therefore not “complete PRISMA and submit.” It is to establish consistency across the entire publication package and then use PRISMA 2020, relevant extensions and journal requirements as independent lenses through which that package is checked.13
A practical pre-submission audit
Confirm the reporting framework
Start with PRISMA 2020, then determine whether the review type, synthesis method or evidence domain requires an official extension or another complementary reporting guideline. Check the current official register rather than relying on a saved checklist from an earlier project.14
Reconcile protocol and completed review
Compare the final question, eligibility criteria, outcomes and planned analyses with the protocol or registration. Record material amendments and explain why they occurred. Do not silently rewrite the history of the review to make the process appear perfectly linear.45
Match every major result to its method
Confirm that each synthesis, subgroup analysis, sensitivity analysis and certainty assessment has a corresponding Methods description. Planned methods that were not used should not simply disappear; where consequential, the report should explain why they were not performed.
Reconcile every numerical surface
Cross-check the manuscript, forest plots, evidence tables, figures, appendices and abstract after the final analysis files are frozen. A corrected analysis is not fully corrected until every downstream number and interpretation has been updated.
Calibrate interpretation to the evidence
Read the Discussion and conclusion against the effect estimates, precision, study limitations, risk-of-bias assessments and certainty judgments. Remove language that implies more certainty, generalizability or causality than the evidence can support.9
Verify the distributed research package
Open every supplementary file, repository and external record from the perspective of a reader. Confirm that links work, labels are intelligible, versions correspond to the submitted manuscript and availability statements describe what is genuinely accessible.6
Audit the abstract independently
The abstract is not a miniature introduction. It is a compressed report of the review. Check it against the current PRISMA 2020 for Abstracts checklist and then compare every claim, number and limitation with the full manuscript.1
Add journal and editorial requirements
Complete the journal’s instructions, authorship and contributorship requirements, conflicts-of-interest declarations, data policies and AI-disclosure rules separately from the PRISMA audit. A journal policy may overlap with PRISMA without being part of PRISMA.13
Methodological boundary. A completed reporting audit demonstrates that required information has been located and communicated. It does not certify that the review’s methods were appropriate, that its included evidence is unbiased, that its synthesis is statistically correct, or that its conclusions are trustworthy. Those judgments require the relevant conduct standards, risk-of-bias assessments, statistical checks and certainty-of-evidence methods.
Frequently asked questions
Is PRISMA 2020 a quality-assessment tool for systematic reviews?
No. PRISMA 2020 is a reporting guideline. It helps authors report why a review was conducted, what methods were used and what was found. A transparently reported review can still contain weak methods, so methodological quality or risk of bias must be evaluated separately using appropriate appraisal methods. Reporting and appraisal are connected because appraisal becomes difficult when important methods are not reported.13
Does a difference between the protocol and final review mean that the review is biased?
Not by itself. Systematic reviews can require legitimate methodological amendments after the protocol is written. The reporting issue is whether material changes are visible, dated where possible and explained sufficiently for readers to understand why they occurred. Unexplained changes are more problematic because readers cannot determine whether the decision was made for a defensible methodological reason or after knowledge of emerging results.35
Does all PRISMA information have to appear in the main manuscript?
No. Reporting may be distributed across the main article, tables, figures, supplementary files, protocol, registration record and repositories. The relevant information must nevertheless be findable and sufficiently cross-referenced. Authors should also follow the target journal’s requirements because journals differ in what must remain in the main manuscript and what may be placed elsewhere.12
Does PRISMA 2020 require authors to make their data and analytic code public?
PRISMA 2020 asks authors to report which materials are publicly available and where they can be found, including extracted data, analysis data and analytic code. That reporting requirement should not be rewritten as a universal mandate that every review must publicly release every research object. Data-sharing obligations can also arise from journal, funder, institutional or disciplinary policies.113
Which PRISMA extension should be used with PRISMA 2020?
The answer depends on the review question and methods. Official extensions address areas such as searches, scoping reviews, network meta-analysis, harms, diagnostic accuracy, living systematic reviews, individual participant data and particular populations or measurement questions. Because extensions are periodically added or updated, authors should check the official PRISMA extension register during manuscript preparation rather than rely on an old list.14
Does completing every PRISMA item guarantee that another team can reproduce the review?
No. Complete reporting improves the information available for checking and reproduction, but reproducibility also depends on the availability and accuracy of the underlying data, code, software details and other research materials. Empirical reproduction studies show that some meta-analyses can be reproduced closely when usable inputs are available, while missing or ambiguous inputs can prevent independent reproduction.67
Should PRISMA adherence be reported as a percentage or quality score?
PRISMA was developed as a reporting checklist, not as a calibrated measure of methodological quality. Percentage or summed scores used in meta-research are study-specific ways of operationalizing reporting completeness and should not be interpreted as validated measures of review quality. For manuscript preparation, item-level mapping is more informative: identify what each applicable item asks for, where it is reported and whether any reporting gap remains.
A publication-ready review should leave a recoverable trail
Transparent reporting does not make a systematic review correct. It makes the review inspectable. That distinction is central. Readers should be able to see where the question came from, which methods were prespecified, which methods were actually used, what changed, which studies contributed to each finding, how the findings were analysed, how uncertainty was handled, and why the final conclusions use the strength of language that they do.
The reporting resources in this collection are therefore best used as components of one control system rather than as isolated downloads. As new templates and methodological guides are added, the collection can expand without changing its governing purpose: preserve the chain from review plan to published evidence so that another person can follow it.
Get the Resource Infrastructure Free Forever
One sign-up grants lifetime access to every template, workbook, and methodology guide on the MetaSyn Resource Infrastructure, current and future. As new resources release, you receive them by email automatically. No spam, no marketing noise, methodology updates from a working meta-analyst, only when there’s something real to share.
References
- Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. https://doi.org/10.1136/bmj.n71
- Page MJ, Moher D, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. PRISMA 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ. 2021;372:n160. https://doi.org/10.1136/bmj.n160
- Cumpston M, Lasserson T, Flemyng E, Page MJ. Chapter III: Reporting the review [updated July 2026]. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 6.5. Cochrane; 2026. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-iii
- Hu K, Zhao L, Zhou Q, Mei F, Gao Q, Chen F, et al. Inconsistencies in study eligibility criteria are common between non-Cochrane systematic reviews and their protocols registered in PROSPERO. Res Synth Methods. 2021;12(3):394-405. https://doi.org/10.1002/jrsm.1476
- Siebert M, Caquelin L, Madera M, Acosta-Dighero R, Naudet F, Roqué M. Assessing the magnitude of changes from protocol to publication: a survey on Cochrane and non-Cochrane systematic reviews. PeerJ. 2023;11:e16016. https://doi.org/10.7717/peerj.16016
- Page MJ, Nguyen PY, Hamilton DG, Haddaway NR, Kanukula R, Moher D, McKenzie JE. Data and code availability statements in systematic reviews of interventions were often missing or inaccurate: a content analysis. J Clin Epidemiol. 2022;147:1-10. https://doi.org/10.1016/j.jclinepi.2022.03.003
- Nguyen PY, McKenzie JE, Alqaidoom Z, Hamilton DG, Moher D, Page MJ. Reproducibility of meta-analytic results in systematic reviews of interventions: meta-research study. BMJ Med. 2025;4:e002024. https://doi.org/10.1136/bmjmed-2025-002024
- Hamilton DG, McKenzie JE, Nguyen PY, Rethlefsen ML, McDonald S, Brennan SE, et al. Evaluation of the replicability of systematic reviews with meta-analyses of the effects of health interventions. Res Synth Methods. 2026;17(3):538-556. https://doi.org/10.1017/rsm.2025.10064
- Yavchitz A, Ravaud P, Altman DG, Moher D, Hróbjartsson A, Lasserson T, Boutron I. A new classification of spin in systematic reviews and meta-analyses was developed and ranked according to the severity. J Clin Epidemiol. 2016;75:56-65. https://doi.org/10.1016/j.jclinepi.2016.01.020
- Rethlefsen ML, Kirtley S, Waffenschmidt S, Ayala AP, Moher D, Page MJ, Koffel JB; PRISMA-S Group. PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews. Syst Rev. 2021;10(1):39. https://doi.org/10.1186/s13643-020-01542-z
- Akl EA, Khabsa J, Iannizzi C, Piechotta V, Kahale LA, Barker JM, et al. Extension of the PRISMA 2020 statement for living systematic reviews (PRISMA-LSR): checklist and explanation. BMJ. 2024;387:e079183. https://doi.org/10.1136/bmj-2024-079183
- Baba A, Farid-Kapadia M, Smith M, Hartling L, Moher D, Hooft L, Offringa M. PRISMA-Children and Adolescents (PRISMA-C) 2026 extension statement and explanation: enhancing the reporting and utility of systematic reviews of interventions in paediatrics. BMJ. 2026;393:e088561. https://doi.org/10.1136/bmj-2025-088561
- International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. Updated January 2026. https://www.icmje.org/recommendations/
- PRISMA Executive. PRISMA extensions: published extensions and extensions under development. Accessed 8 August 2026. https://www.prisma-statement.org/extensions
Scope note. PRISMA 2020 governs reporting, not the methodological conduct or critical appraisal of a systematic review. Review authors should use the methodological guidance appropriate to their review design and consult the current PRISMA extension register because reporting guidance continues to evolve.