Contents
On this page
- Definition and alternative terms
- Patterns of duplication
- Prevalence by specialty and across databases
- Effects on the evidence base
- Permitted forms of secondary publication
- Methods of detection
- Consequences of detection
- The footnote that legitimises a secondary publication
- Assessment of publication status before submission
- Key points
- Application to your study
- Frequently asked questions
Academy · Research Publication · Duplicate publication
Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication
Publishing the same data twice is primarily a matter of data integrity rather than of writing conduct, because any pooled analysis of the literature counts the same participants more than once. The harm has been measured: in one documented case, inclusion of the duplicated reports overestimated a drug’s efficacy by 23%. The danger of a duplicate lies not in the overlap but in the absence of a cross-reference.
Most authors who encounter a duplicate publication problem did not intend to deceive. Typically they had a dataset that supported more than one paper, a co-author in another country who wanted a version for a local journal, or a congress abstract that developed into an article. The rules in this area are more permissive than commonly assumed — secondary publication is explicitly allowed under stated conditions — but considerably less tolerant of concealment. This article sets out the distinction, the findings of published audits, and the footnote that legitimises a second publication.
Learning objectives
- What distinguishes duplicate publication from acceptable secondary publication and from text recycling
- How often duplicates appear: 40% of systematic reviews in one audit, and prevalence of 7% to 25% of records across multi-database searches
- What duplication does to a pooled estimate, with a documented case of a 23% overestimate
- Why covert duplicates — those with no cross-reference — are the ones that cause harm
- The conditions under which a second publication is permitted, and the footnote it requires
- How editors and systematic reviewers detect overlap, and what shows up in retraction statistics
Common misconceptions
- “It is the same authors, so it cannot be duplicate publication.” Authorship is largely irrelevant to the harm, and duplicates frequently involve different authors: von Elm and colleagues found that 64% of the duplicates they identified had partly or completely different authorship from the main article.
- “Publishing the same study twice is always prohibited.” This is not the case. Secondary publication — typically in another language, for a different readership — is permitted under the ICMJE provisions, with the agreement of both editors and a footnote citing the primary report.
- “Duplication is a courtesy issue between journals.” Duplication distorts pooled estimates. The case study by Tramer and colleagues showed that including duplicated data overestimated the antiemetic efficacy of ondansetron by 23%.
- “Duplicates are rare and obvious.” A text-similarity scan of 62,213 Medline citations found roughly 1.35% of same-author citation pairs similar enough to count as duplicates, extrapolating to about 117,400 Medline citations, plus or minus 40,000.
- “Editors and authors agree on where the line is.” Yank and Barnes surveyed both groups in the Journal of Medical Ethics and found consensus on some cases and genuine disagreement on others. Where the boundary is unclear, the editor’s judgement prevails, and authors should therefore consult the editor.
Definition and alternative terms
Duplicate publication — also called redundant publication, and when the overlap is partial, salami publication — means publishing the same data, or a substantially overlapping subset of them, more than once without disclosure. The term is long established, and the objection to the practice is older still. Angell and Relman set it out in the New England Journal of Medicine in 1989, and Kassirer and Angell issued a reminder in the same journal six years later 28,29. DeAngelis returned to it in JAMA under the title “Duplicate publication, multiple problems”, and Alfonso, Bermejo and Segovia asked in Revista Espanola de Cardiologia whether the literature could afford it 22,24.
Three practices must be distinguished before this subject can be analysed, because they are governed differently and detected differently.
Duplicate publication reuses the data. The reader who pools the literature double-counts participants. This is a data-integrity problem and the instrument is the ICMJE provisions on overlapping publication 1.
Text recycling reuses your own prose, generally in methods or background, without reusing data. COPE issued text recycling guidelines on it, and Moskovitz, in Science and Engineering Ethics, argues much of it is harmless and sometimes unavoidable 2,8. Anson and Moskovitz’s text-analytic study in Accountability in Research found it common rather than exceptional in published STEM articles 6.
Plagiarism reuses someone else’s words or ideas. Das and Panjabi set out why it matters particularly for medical writers in Perspectives in Clinical Research 15. The three overlap in detection software and nowhere else. For the surrounding framework, see the publication ethics hub; for the third, the consequences of plagiarism.
Patterns of duplication
Von Elm and colleagues’ analysis in JAMA is the most useful single study here because it classified what it found rather than only counting it 23. They examined the articles used in 141 systematic reviews in anaesthesia and analgesia 23. Authors of 56 of those reviews — 40% — reported identifying duplicate publications, and those reviews covered 1,131 main articles with 103 duplicates arising from 78 main articles.
The patterns were not uniform. Some duplicates reproduced the main article almost entirely. Others divided a single study’s participants across reports, added participants to a previously published cohort without indicating which were new, or reported different outcomes from the same patients in separate papers. The practical consequence is that a reviewer cannot detect duplication by searching for identical papers; participant counts, recruitment windows and centre lists must be reconciled across reports, which is slow and often inconclusive.
Two features of their duplicates stand out. Covert duplicates — those carrying no cross-reference to the main article — made up 5.3% of articles, 65 of 1,234 23. Of the duplicates identified, 33% were industry sponsored and 64% had partly or completely different authorship from the main report, with a median delay of about a year between the main article and the duplicate. A duplicate is therefore not reliably a paper with the same author list, which explains why such duplicates are difficult to detect 23.
Prevalence by specialty and across databases
The prevalence figures vary widely, and the variation is informative rather than noise. Qi and colleagues, in the World Journal of Meta-Analysis, reported that across their own systematic reviews searching three or more databases, the prevalence of duplicate records ranged from 7% to 25% 11. They also summarised specialty audits: about 14% of 660 surgical original articles showed some potential redundancy, under 1% of plastic surgery articles did, and 10% — 184 of 1,914 — of Chinese Budd-Chiari syndrome articles were covert duplicates 10,11. Schein and Paladugu had raised the surgical question in Surgery, Durani examined the plastic surgery literature in the Journal of Plastic, Reconstructive and Aesthetic Surgery, and Arrive and colleagues audited redundancy in Radiology 19,21,26.
At the whole-database scale, Errami and colleagues ran a text-similarity scan of 62,213 Medline citations and estimated that about 1.35% of citations sharing authors were similar enough to count as duplicates, which extrapolates to roughly 117,400 Medline citations, plus or minus 40,000 20,23. Sun and colleagues characterised text similarity in full-text biomedical publications in PLoS ONE on the same general approach.
The range should be interpreted as a whole rather than used as a source from which to select a single figure. A 25-fold spread between specialties almost certainly reflects differences in what was counted and in how thoroughly duplicates were sought, not a genuine 25-fold difference in conduct. All the audits agree that the rate is not negligible in any field examined to date.
Prevalence estimates from published audits
| Dataset | What was counted | Finding |
|---|---|---|
| 141 systematic reviews in anaesthesia and analgesia von Elm et al., JAMA | Reviews whose authors identified duplicates among their included articles | 56 of 141 reviews (40%); 103 duplicates from 78 main articles, across 1,131 main articles |
| The same article set von Elm et al., JAMA | Covert duplicates — no cross-reference to the main article | 5.3% of articles (65 of 1,234); 33% industry sponsored; 64% with partly or completely different authorship |
| Multi-database systematic reviews Qi et al., World Journal of Meta-Analysis | Duplicate records in searches of three or more databases | Prevalence ranged from 7% to 25% |
| 660 surgical original articles summarised by Qi et al. | Articles showing some potential redundancy | About 14% |
| Plastic surgery literature summarised by Qi et al. | Articles showing redundancy | Under 1% |
| 1,914 Chinese Budd-Chiari syndrome articles summarised by Qi et al. | Covert duplicates | 10% (184 of 1,914) |
| 62,213 Medline citations Errami et al., Bioinformatics | Same-author citation pairs similar enough to count as duplicates | About 1.35%, extrapolating to roughly 117,400 Medline citations (±40,000) |
Figures as reported in each source. The definitions of duplication, the fields and the detection methods differ between studies, so the rows are not directly comparable and should not be pooled; the spread between specialties very likely reflects how thoroughly duplicates were sought.
Effects on the evidence base
The clearest demonstration of harm is a single case study, and it merits detailed consideration because it is the most persuasive argument. Tramer and colleagues, reported in BMJ and summarised by Qi and colleagues, examined the trials of ondansetron for postoperative emesis and found 17% of randomised controlled trials and 28% of patient data duplicated 11,27. Including the duplicated data overestimated the drug’s antiemetic efficacy by 23%.
This single figure encapsulates the objection. A meta-analysis weights studies by size; duplicated participants inflate the weight of whichever result happened to be published twice, and the results published twice are not a random sample. A reviewer cannot correct for duplication that is not visible.
The burden falls on those conducting the pooling, and Qi and colleagues have written about it from that perspective: their work in PLoS ONE on finding duplicates among PubMed, EMBASE and the Cochrane Library addresses the mechanics, and their World Journal of Meta-Analysis paper argues the issue is critical but routinely neglected in review methods 11,12. Anyone running a systematic review or a meta-analysis should treat duplicate identification as a named step with a recorded method, not as something that falls out of deduplication software.
The ondansetron case study, as reported by Tramer MR, Reynolds DJM, Moore RA, McQuay HJ in BMJ and summarised by Qi XS, Bai M, Yang ZP, Ren WR in the World Journal of Meta-Analysis. A single demonstration, not a general effect size.
Permitted forms of secondary publication
The rules are more accommodating than most authors assume, and awareness of the permissions matters because they remove the incentive to conceal.
Secondary publication is allowed under the ICMJE provisions, most commonly to reach a different readership in another language. The conditions are specific: the editors of both journals agree, the secondary version reflects the same data and interpretation as the primary, it appears at an interval the primary editor has accepted, it is aimed at a different group of readers, and it carries a footnote telling readers that the paper was published in whole or in part elsewhere, with a citation to the primary report 1. These conditions are not onerous; the most common failure is omission of the footnote.
Reporting different outcomes from one study in separate papers can be legitimate where the papers address genuinely different questions and each cites the others and describes the shared cohort. What makes it salami publication rather than a legitimate series is the absence of those cross-references and a participant count a reader cannot reconcile.
Abstracts and presentations are generally treated as prior presentation rather than prior publication, but the decision rests with the editor, and they should therefore be disclosed in the cover letter by citation.
Recycled methods text is a separate question governed by COPE’s text recycling guidelines and the journal’s own policy 2,4. The reuse should be disclosed and the earlier paper cited. Detection software will flag it in either case: Foltynek, Meuschke and Gipp’s review in ACM Computing Surveys describes methods that identify overlap regardless of whose text it is 7.
Permitted, conditional, and not permitted
| What you want to do | Status | What it requires |
|---|---|---|
| Publish a translated version for a different readership | Permitted | Both editors agree; same data and interpretation; an interval the primary editor accepts; a footnote stating prior publication and citing the primary report |
| Publish a full paper after a congress abstract | Normally permitted | Disclose the abstract by citation in the cover letter, cite it where the overlapping results appear, and state what the full paper adds |
| Report different outcomes from one cohort in separate papers | Conditional | Genuinely different questions; each paper cites the others; the shared cohort and participant counts are reconcilable by a reader |
| Reuse your own methods text in a new paper | Conditional — and contested | Disclose the reuse, cite the earlier paper, and follow the journal’s policy; detection software will flag it either way |
| Add new participants to a previously published cohort | Conditional | The earlier report must be cited and the new participants identified, so that nobody pooling the literature counts anyone twice |
| Publish the same data in a second journal without telling either | Not permitted | Nothing makes this acceptable; it is the covert pattern that the audits associate with harm and with retraction |
Conditions as set out in the ICMJE Recommendations and COPE’s text recycling guidelines. Where a journal’s own policy is stricter, it governs.
Methods of detection
Detection happens in three places, and only the first is automated. At the journal, similarity software compares the submission against the published literature. Zhang, in Learned Publishing, described how this works in editorial practice; Sun and colleagues and Errami and colleagues developed the underlying text-similarity approaches, and Errami’s Déjà vu work in Bioinformatics remains the reference point for the scale of the problem in Medline 16,17,20.
At the systematic review, duplicates are found by hand, because the software finds duplicate records of the same article rather than duplicate publications of the same study. Qi and colleagues describe the reconciliation work in PLoS ONE: matching participant numbers, recruitment dates, centre lists, baseline tables and funding statements across candidate reports 12. This is the step that identifies covert duplicates, and it is also the step most often omitted.
Detection by the reader occurs by chance. Several of the specialty audits began when someone noticed two similar-looking papers. For a prospective check of a manuscript before an editor conducts one, the relevant service is a plagiarism check; how an editor or reviewer reasons from a similarity report is covered in identification of multiple publication.
Consequences of detection
Retraction is a real outcome rather than a theoretical one, and its share is measurable. Fang, Steen and Casadevall, in the Proceedings of the National Academy of Sciences of the USA, classified 2,047 retracted biomedical and life-science articles indexed in PubMed as of 3 May 2012 14. Misconduct accounted for 67.4% of retractions, and within that, duplicate publication accounted for 14.2% — against 43.4% for fraud or suspected fraud, 9.8% for plagiarism, and 21.3% for error. Duplicate publication is therefore the third most common reason a biomedical paper is withdrawn.
Short of retraction, the usual outcomes are a correction carrying the cross-reference that should have been present, an editorial note, a rejection with a letter to the author’s institution, or an informal decision not to consider future submissions. Which of these follows is not predictable from the facts, partly because editors and authors do not agree on the severity of particular cases: Yank and Barnes’ cross-sectional survey in the Journal of Medical Ethics found consensus on some scenarios and substantial disagreement on others 25. That disagreement is the strongest practical argument for consulting the editor in advance, because the editor’s position cannot be inferred from the guidance.
The footnote that legitimises a secondary publication
In the published article, almost all of the foregoing reduces to a few lines. The two versions below concern the same paper: a Spanish-language version of a trial report originally published in English, prepared for clinicians who would not otherwise read it, with the agreement of both editors.
A secondary publication, with and without the footnote that legitimises it
A covert duplicate
Nota del autor: Parte de estos resultados se presentó previamente en un congreso internacional. [Author note: Some of these results were previously presented at an international congress.]
No reader can find the primary report. There is no citation, so the paper is unlinkable and a systematic reviewer who retrieves both has no way to know they are one study.
“Some of these results” conceals the scale of the overlap. This is the whole trial, not a subset, and the wording suggests otherwise.
The congress presentation is mentioned; the English-language paper is not. That is a disclosure of the lesser prior appearance and silence on the greater one.
Nothing records the editors’ agreement. Without it the secondary publication does not meet the ICMJE conditions however the data are described.
No statement that the data are unchanged. A reader cannot tell whether this is the same analysis or a reanalysis, which is the difference between a secondary publication and a second study.
A disclosed secondary publication
This article is a secondary publication. It was published in full in English as: Author AB, Author CD, Author EF, et al. Efficacy and safety of [intervention] in [population]: a randomised, double-blind, placebo-controlled trial. Journal Name. 2024;12(4):301–312. doi:10.0000/example. It is republished here in Spanish, with the agreement of the editors of both journals, for readers in clinical practice who do not routinely read the English-language literature. The data, analyses and interpretation are those of the primary report and are unchanged; no additional participants, outcomes or analyses are included. The trial is registered as NCT00000000.
The primary report is cited in full with a DOI, so the two articles are linkable and the duplication is visible to anyone pooling the literature.
The purpose and the distinct readership are stated, which is one of the ICMJE conditions rather than a courtesy.
Both editors’ agreement is on the record.
The paper states positively that no participants, outcomes or analyses have been added — the sentence that prevents a reviewer treating it as independent evidence.
The registration number ties both reports to the same trial, which is the most reliable single cross-reference available.
The second version is permitted and the first is a covert duplicate, and the only difference is disclosure. That is the pattern the audits keep finding: von Elm and colleagues identified covert duplicates — those with no cross-reference — in 5.3% of the articles they examined, and it is the concealment rather than the republication that does the damage and attracts the sanction.
Assessment of publication status before submission
Two questions resolve almost every practical case. The first is whether a reader pooling this literature could count any participant twice without knowing it; if so, a cross-reference is required, wherever else the paper is submitted. The second is whether any public appearance of these data — abstract, poster, preprint, registry posting, earlier paper from the same cohort, foreign-language version — has not been disclosed to the editor; if so, it should be disclosed by citation in the cover letter.
Neither question requires a judgement about intent, and this is deliberate. Von Elm and colleagues’ finding that the damaging duplicates were the covert ones, and Fang and colleagues’ finding that duplicate publication accounts for 14.2% of retractions, point in the same direction: the offence that is sanctioned is concealment 14. Disclosure requires only a sentence. Where it is genuinely uncertain whether a second paper is permitted — and the editor and author surveys suggest that reasonable people differ — authors should write to the editor describing both papers and retain the reply.
Key points
Key points
Duplicate publication harms the evidence base by allowing the same participants to be counted twice, and the measured consequence can be substantial: in the ondansetron case, inclusion of duplicated data overestimated efficacy by 23%. Duplication is not rare: 40% of the systematic reviews in one audit reported finding duplicates, the prevalence of duplicate records across multi-database searches ranged from 7% to 25%, and duplication accounts for 14.2% of retractions in the largest published classification. The rules prohibit concealment rather than republication. Secondary publication is permitted with the agreement of both editors, unchanged data and a footnote citing the primary report. Covert duplicates, which carry no cross-reference, are those that cause harm and those that are retracted.
Application to your study
Application to your study
- Address the double-counting question first. The question is whether a reader pooling this literature could count any participant twice without knowing it; if so, a cross-reference is required regardless of other considerations.
- List every prior public appearance of the data before writing the cover letter — congress abstract, poster, preprint, registry results posting, earlier paper from the same cohort, foreign-language version — and disclose each by citation.
- Cite the earlier report at the point where the overlapping results appear, not only in the cover letter, since the cover letter is read once whereas the article is read for years.
- When adding participants to a published cohort, state how many are new and cite the earlier report; this is the most common form of unintentional duplication.
- For a secondary publication, obtain the written agreement of both editors and give the full citation in a footnote, with a sentence stating that no participants, outcomes or analyses have been added.
- Use the registration number as the cross-reference. It survives title changes, translation and differing author lists better than anything else.
- In a systematic review, make duplicate identification a named step with a recorded method: reconcile participant counts, recruitment windows, centre lists and funding statements, because deduplication software finds duplicate records, not duplicate studies.
- Where it is unclear whether a second paper is permitted, write to the editor and retain the reply. Editors and authors demonstrably disagree about particular cases, so the answer cannot be inferred from the guidance alone.
Frequently asked questions
They reuse different things and are governed differently. Duplicate publication reuses the data: the same participants appear in more than one report, so anyone pooling the literature double-counts them. That is a data-integrity problem, governed by the ICMJE provisions on overlapping publication. Plagiarism reuses another person’s words or ideas as the author’s own, which is an attribution problem. Between the two sits text recycling — the reuse of an author’s own prose without reuse of data — which COPE addresses in its text recycling guidelines and which Moskovitz, in Science and Engineering Ethics, argues is frequently harmless. All three are flagged by the same similarity software, which explains why they are frequently confused.
Yes, under stated conditions. The ICMJE provisions permit secondary publication, most often in another language to reach a readership that would not otherwise see the work. The conditions are that the editors of both journals agree, that the secondary version reflects the same data and interpretation as the primary, that it appears after an interval the primary editor accepts, that it is aimed at a different group of readers, and that it carries a footnote stating that the article has been published in whole or in part elsewhere, with a full citation to the primary report. Authors rarely fail to meet the first four conditions, but they frequently omit the footnote, and this omission converts a permitted secondary publication into a covert duplicate.
Sufficiently common that every audit that sought it found some instances, with wide variation. Von Elm and colleagues, in JAMA, examined the articles used in 141 anaesthesia and analgesia systematic reviews and found that authors of 56 — 40% — had identified duplicates; those reviews covered 1,131 main articles with 103 duplicates from 78 main articles. Qi and colleagues reported duplicate record prevalence of 7% to 25% across their own multi-database reviews, and summarised specialty audits finding about 14% potential redundancy in 660 surgical articles, under 1% in plastic surgery, and 10% covert duplicates among 1,914 Chinese Budd-Chiari syndrome articles. The 25-fold spread probably reflects differences in how thoroughly each group searched.
It changes results, and there is a documented case. Tramer and colleagues examined the trials of ondansetron for postoperative emesis, reported in BMJ and summarised by Qi and colleagues: 17% of randomised controlled trials and 28% of patient data were duplicated, and including the duplicated data overestimated the drug’s antiemetic efficacy by 23%. The mechanism is straightforward — a meta-analysis weights studies by size, so duplicated participants inflate the weight of whichever result happened to be published twice, and results published twice are not a random sample. The 23% figure should be treated as one well-documented case rather than a general effect size, although the direction of the problem is not in doubt.
A covert duplicate is a second publication of the same data carrying no cross-reference to the main article, so a reader has no means of connecting them. Von Elm and colleagues found covert duplicates in 5.3% of the articles they examined, 65 of 1,234. They are more harmful for a straightforward reason: a disclosed second publication can be identified and excluded by anyone pooling the literature, whereas a covert one cannot. Two further findings explain why they are difficult to detect — 33% of the duplicates were industry sponsored and 64% had partly or completely different authorship from the main report, with a median delay of about a year. A search for the same author list will not identify them.
The outcomes range from a correction to a retraction, and which one follows is not reliably predictable. Fang, Steen and Casadevall, classifying 2,047 retracted biomedical and life-science articles in the Proceedings of the National Academy of Sciences of the USA, found that duplicate publication accounted for 14.2% of retractions — the third most common reason, after fraud or suspected fraud at 43.4% and ahead of plagiarism at 9.8%; error accounted for 21.3%. Short of retraction, the observed outcomes include a correction adding the missing cross-reference, an editorial note, a rejection with a letter to the author’s institution, and an informal decision not to consider future submissions. Yank and Barnes found that editors and authors genuinely disagree about the seriousness of particular cases.
Manually, and it is the step most often omitted. Deduplication software finds duplicate records of the same article across databases, which is a different problem from duplicate publications of the same study. Qi and colleagues describe the reconciliation work in PLoS ONE and in the World Journal of Meta-Analysis: comparing participant numbers, recruitment windows, centre lists, baseline characteristic tables, funding statements and registration numbers across candidate reports until every participant can be accounted for once. Registration numbers are the single most useful field, because they survive translation, title changes and differing author lists. The step should be built into the protocol with a stated method rather than treated as part of screening.
Legitimacy depends on whether a reader can reconcile the papers, not on how many papers there are. Separate papers from one cohort are legitimate where each addresses a genuinely different question, each cites the others, and the shared cohort and participant counts are described clearly enough that nobody pools them twice. The practice becomes salami publication when those cross-references are absent, when the division is driven by publication count rather than by distinct questions, or when a reader cannot tell whether the 400 participants in paper two include the 250 in paper one. A practical test is to write the cohort description and the cross-references first; if they are difficult to write transparently, the division is probably the problem.
References
- International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. ICMJE. 2026. read 9 October 2026. icmje.org
- Committee on Publication Ethics. Text recycling guidelines. COPE. 2013. read 9 October 2026. par.nsf.gov
- International Committee of Medical Journal Editors. The New ICMJE Recommendations (August 2013). ICMJE. 2013. read 9 October 2026. icmje.org
- Committee on Publication Ethics. Core Practices. COPE. read 9 October 2026. equator-network.org
- DeTora LM, Toroser D, Sykes A, Vanderlinden C, Plunkett FJ, Lane T, Hanekamp E, Dormer L, DiBiasi F, Bridges D, Baltzer L, Citrome L. Good Publication Practice (GPP) Guidelines for Company-Sponsored Biomedical Research: 2022 Update. Annals of Internal Medicine. 2022;175(9):1298-1304. doi:10.7326/M22-1460 read 9 October 2026. doi.org/10.7326/M22-1460
- Anson IG, Moskovitz C. Text recycling in STEM: a text-analytic study of recently published research articles. Accountability in Research. 2020. doi:10.1080/08989621.2020.1850284 read 9 October 2026. doi.org/10.1080/08989621.2020.1850284
- Foltynek T, Meuschke N, Gipp B. Academic plagiarism detection: a systematic literature review. ACM Computing Surveys. 2019;52(6):1-42. doi:10.1145/3345317 read 9 October 2026. doi.org/10.1145/3345317
- Moskovitz C. Text recycling in scientific writing. Science and Engineering Ethics. 2019;25(3):813-851. doi:10.1007/s11948-017-0008-y read 9 October 2026. doi.org/10.1007/s11948-017-0008-y
- Battisti WP, Wager E, Baltzer L, Bridges D, Cairns A, Carswell CI, Citrome L, Gurr JA, Mooney LA, Moore BJ, Pena T, Sanes-Miller CH, Veitch K, Woolley KL, Yarker YE. Good Publication Practice for Communicating Company-Sponsored Medical Research: GPP3. Annals of Internal Medicine. 2015;163(6):461-464. doi:10.7326/M15-0288 read 9 October 2026. doi.org/10.7326/M15-0288
- Qi X, Deng H, Guo X, Fan D. Covert duplicate publications in Budd-Chiari syndrome articles in China. The American Journal of Medicine. 2013;126:633-639.e2. read 9 October 2026. wjgnet.com
- Qi XS, Bai M, Yang ZP, Ren WR. Duplicates in systematic reviews: a critical, but often neglected issue. World Journal of Meta-Analysis. 2013;1(3):97-101. doi:10.13105/wjma.v1.i3.97 read 9 October 2026. doi.org/10.13105/wjma.v1.i3.97
- Qi X, Yang M, Ren W, Jia J, Wang J, Han G, Fan D. Find duplicates among the PubMed, EMBASE, and Cochrane Library databases in systematic review. PLoS ONE. 2013;8:e71838. doi:10.1371/journal.pone.0071838 read 9 October 2026. doi.org/10.1371/journal.pone.0071838
- Turner L, Shamseer L, Altman DG, Schulz KF, Moher D. Does use of the CONSORT statement impact the completeness of reporting of randomised controlled trials published in medical journals? A Cochrane review. Systematic Reviews. 2012;1:60. read 9 October 2026. care-statement.org
- Fang FC, Steen RG, Casadevall A. Misconduct accounts for the majority of retracted scientific publications. Proceedings of the National Academy of Sciences of the USA. 2012;109(42):17028-17033. doi:10.1073/pnas.1212247109 read 9 October 2026. doi.org/10.1073/pnas.1212247109
- Das N, Panjabi M. Plagiarism: why is it such a big issue for medical writers? Perspectives in Clinical Research. 2011;2(2):67-71. doi:10.4103/2229-3485.80370 read 9 October 2026. doi.org/10.4103/2229-3485.80370
- Zhang H. CrossCheck: an effective tool for detecting plagiarism. Learned Publishing. 2010;23:9-14. read 9 October 2026. pmc.ncbi.nlm.nih.gov
- Sun Z, Errami M, Long T, Renard C, Choradia N, Garner H. Systematic characterizations of text similarity in full text biomedical publications. PLoS ONE. 2010;5(9):e12704. doi:10.1371/journal.pone.0012704 read 9 October 2026. doi.org/10.1371/journal.pone.0012704
- Graf C, Battisti WP, Bridges D, Bruce-Winkler V, Conaty JM, Ellison JM, Field EA, Gurr JA, Marx M-E, Patel M, Sanes-Miller C, Yarker YE. Good publication practice for communicating company sponsored medical research: the GPP2 guidelines. BMJ. 2009;339:b4330. doi:10.1136/bmj.b4330 read 9 October 2026. doi.org/10.1136/bmj.b4330
- Arrive L, Lewin M, Dono P, Monnier-Cholley L, Hoeffel C, Tubiana JM. Redundant publication in the journal Radiology. Radiology. 2008;247:836-840. doi:10.1148/radiol.2473070969 read 9 October 2026. doi.org/10.1148/radiol.2473070969
- Errami M, Hicks JM, Fisher W, Trusty D, Wren JD, Long TC, Garner HR. Deja vu - a study of duplicate citations in Medline. Bioinformatics. 2007. read 9 October 2026. ori.hhs.gov
- Durani P. Duplicate publications: redundancy in plastic surgery literature. Journal of Plastic, Reconstructive & Aesthetic Surgery. 2006;59:975-977. read 9 October 2026. wjgnet.com
- Alfonso F, Bermejo J, Segovia J. Duplicate or redundant publication: can we afford it? Revista Espanola de Cardiologia. 2005;58(5):601-604. doi:10.1016/S1885-5857(06)60739-1 read 9 October 2026. doi.org/10.1016/S1885-5857(06)60739-1
- von Elm E, Poglia G, Walder B, Tramer MR. Different patterns of duplicate publication: an analysis of articles used in systematic reviews. JAMA. 2004;291(8):974-980. doi:10.1001/jama.291.8.974 read 9 October 2026. doi.org/10.1001/jama.291.8.974
- DeAngelis CD. Duplicate publication, multiple problems. JAMA. 2004;292:1745-1746. doi:10.1001/jama.292.14.1745 read 9 October 2026. doi.org/10.1001/jama.292.14.1745
- Yank V, Barnes D. Consensus and contention regarding redundant publications in clinical research: cross-sectional survey of editors and authors. Journal of Medical Ethics. 2003;29:109-114. read 9 October 2026. revespcardiol.org
- Schein M, Paladugu R. Redundant surgical publications: tip of the iceberg? Surgery. 2001;129:655-661. read 9 October 2026. wjgnet.com
- Tramer MR, Reynolds DJM, Moore RA, McQuay HJ. Impact of covert duplicate publication on meta-analysis: a case study. BMJ. 1997;315:635-640. read 9 October 2026. wjgnet.com
- Kassirer JP, Angell M. Redundant publication: a reminder. New England Journal of Medicine. 1995;333:449-450. doi:10.1056/NEJM199508173330709 read 9 October 2026. doi.org/10.1056/NEJM199508173330709
- Angell M, Relman AS. Redundant publication. New England Journal of Medicine. 1989;320:1212-1214. doi:10.1056/NEJM198905043201812 read 9 October 2026. doi.org/10.1056/NEJM198905043201812
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 9). Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication. Pubrica. https://pubrica.com/academy/concepts-definitions/duplication-publication/
Pubrica Academy. “Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication.” Pubrica, 9 Oct. 2026, pubrica.com/academy/concepts-definitions/duplication-publication/.
Pubrica Academy (2026) Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication. Pubrica. Available at: https://pubrica.com/academy/concepts-definitions/duplication-publication/ (Accessed: 15 October 2026).
Pubrica Academy. “Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication.” Pubrica. October 9, 2026. https://pubrica.com/academy/concepts-definitions/duplication-publication/
Pubrica Academy, ‘Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication’ (Pubrica, 9 October 2026) <https://pubrica.com/academy/concepts-definitions/duplication-publication/> accessed 15 October 2026.
Pubrica Academy. Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication. [Internet]. Pubrica; 2026 Oct 9 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/concepts-definitions/duplication-publication/
Pubrica Academy, “Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication,” Pubrica, Oct. 9, 2026. [Online]. Available: https://pubrica.com/academy/concepts-definitions/duplication-publication/
@misc{duplication_publication_2026,
author = {{Pubrica Academy}},
title = {Duplicate Publication: Definitions, Effects on the Evidence Base and Permitted Secondary Publication},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/concepts-definitions/duplication-publication/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Prior-appearance disclosure worksheet
Complete this before writing the cover letter, for every manuscript reporting data that have existed for more than a few weeks. The point is to produce the disclosure text, not to decide whether disclosure is needed — it always is.
PRIOR-APPEARANCE DISCLOSURE WORKSHEET
Manuscript: ______________________ Target journal: ____________
Study / cohort: ______________________ Registration no: ____________
1. EVERY PUBLIC APPEARANCE OF THESE DATA
For each, give a full citation or a precise description plus a date.
Congress abstract(s) ............ ______________________________
Poster / oral presentation ...... ______________________________
Late-breaker ..................... ______________________________
Press release .................... ______________________________
Preprint (server + DOI) ......... ______________________________
Registry results posting ........ ______________________________
Thesis or dissertation chapter .. ______________________________
Earlier paper, same cohort ...... ______________________________
Foreign-language version ........ ______________________________
Report to a regulator, if public . ______________________________
2. THE DOUBLE-COUNTING TEST
Total participants in this manuscript: ____________
Of those, how many appeared in any earlier report? ____________
Which earlier report(s): ______________________________________
[ ] The manuscript states the overlap explicitly, with a citation, at the
point where the overlapping results appear
[ ] A reader could account for every participant exactly once across all
our published reports
3. IF THIS IS A SECONDARY PUBLICATION
Primary report, full citation: ________________________________
[ ] Primary journal's editor has agreed, in writing (date: ________)
[ ] This journal's editor has agreed, in writing (date: ________)
[ ] Data, analyses and interpretation are unchanged
[ ] Intended readership differs, and the footnote says how
[ ] Footnote drafted, containing: statement of prior publication; full
citation with DOI; the agreement of both editors; a sentence confirming
no participants, outcomes or analyses added; registration number
4. IF THIS IS ONE OF SEVERAL PAPERS FROM ONE COHORT
The distinct question this paper answers: ______________________
Companion papers, cited in this manuscript: ____________________
[ ] Each companion paper is cited here
[ ] The cohort description is identical across the papers
[ ] Participant counts across the set are reconcilable
5. RECYCLED TEXT
Sections reusing our own earlier wording: ______________________
Source paper(s): ______________________________________________
[ ] Source cited
[ ] Reuse disclosed in the cover letter
[ ] Target journal's text recycling policy read
6. COVER LETTER PARAGRAPH (draft it here)
______________________________________________________________
______________________________________________________________
7. UNRESOLVED QUESTION FOR THE EDITOR
______________________________________________________________
Date asked: ____________ Reply received and filed: Y / N Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
How Pubrica can help
Pubrica’s publication-support and review teams undertake two distinct tasks in this area. For authors, every prior appearance of a dataset is mapped, the cross-references and secondary-publication footnote are drafted, and any planned second paper likely to be regarded as salami publication is identified. For systematic reviewers, duplicate identification is conducted as a recorded step, reconciling participant counts, recruitment windows, centre lists and registration numbers across candidate reports rather than relying on record-level deduplication, which is the step that fails to detect covert duplicates.