Contents
On this page
- The sequence of consequences
- Evidence from the retraction record
- Documented sanctions
- Text recycling as a contested practice
- Duplicate publication and its consequences
- Detection methods and the interpretation of similarity scores
- Inappropriate and appropriate handling of a flagged methods paragraph
- Responding to an allegation or to plagiarism identified in published work
- Effective prevention practices
- Key points
- Application to your study
- Frequently asked questions
Academy · Research Publication · Consequences of plagiarism
The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions
The consequences of plagiarism are not a single event but a sequence that outlasts the paper: an editorial investigation, a retraction notice that remains permanently attached to the record, a report to the author’s institution and, in documented cases, a publication ban, loss of funding or action against a professional licence. Plagiarism accounts for a smaller proportion of retractions than many researchers assume, and the related offence of duplicate publication for a larger one.
Plagiarism is usually taught as a warning and seldom as a sequence of events, which leads authors to misjudge both the likelihood and the nature of the consequences. A paper is not withdrawn without trace. It remains in the literature with a retraction notice attached, discoverable by anyone who searches the author’s name for the remainder of their career, while an institutional process proceeds in parallel on a timescale the author does not control. This article sets out what the published record shows about that sequence, how frequently it is initiated, and where the contested area of reusing one’s own words begins.
Learning objectives
- The escalation sequence from similarity flag to institutional finding, and the party that controls each step
- The proportion of retractions citing plagiarism and duplicate publication, from an analysis of 2,047 retracted articles
- The specific sanctions documented in the literature, including a three-month suspension from medical practice
- The extent of text recycling in practice, measured across 400 articles, and the reasons researchers disagree about it
- The reasons a similarity percentage does not constitute a finding, and the factors editors examine
- Appropriate responses to an allegation, or to the discovery of recycled text in previously published work
Common misconceptions
- “Plagiarism is the main reason papers get retracted.” Fang, Steen and Casadevall, in PNAS, found plagiarism cited in 9.8% of 2,047 retracted articles, against 43.4% for fraud or suspected fraud and 14.2% for duplicate publication.
- “A retracted paper disappears.” A retracted paper remains indexed and citable with a retraction notice attached, which is the purpose of a retraction rather than an incidental effect.
- “A high similarity score means plagiarism.” This is not the case. Foltynek, Meuschke and Gipp’s systematic review in ACM Computing Surveys describes what detection systems can and cannot establish; the judgement is always an editorial one.
- “Reusing my own methods paragraph is obviously fine.” Researchers disagree on this question. Anson and Moskovitz report that Hall and colleagues found 42% of expert and 56% of novice researchers considered recycling text from prior papers inappropriate.
- “Only the journal is involved.” Das and Panjabi, in Perspectives in Clinical Research, document sanctions that have included retraction, publication bans, reports to employers, loss of funding and, in one UK case, a three-month suspension from medical practice.
The sequence of consequences
What follows a plagiarism allegation is a process involving several independent actors, and its most important structural feature is that the author controls almost none of the timing. The order is important, because the decisions that determine the outcome are made in the first two steps, while many authors are still disputing the similarity score.
Detection. Detection usually occurs at submission, through similarity screening, or after publication, through a reader’s report. Zhang, writing in Learned Publishing, described CrossCheck’s introduction as an effective screening tool, and screening is now routine at most journals 17. A flag is not an allegation; it is a prompt for an editor to examine the manuscript.
Editorial assessment. The editor compares the flagged passages against the sources, judges whether the overlap is substantive or incidental and, in accordance with COPE’s Core Practices, normally writes to the corresponding author for an explanation before taking any further step 4. At this stage a candid, specific reply can improve the outcome, whereas a defensive reply tends to worsen it.
Editorial action. Depending on the finding, the outcome may be rejection without further consequence; acceptance after revision where the overlap was procedural and is now attributed; a correction; an expression of concern while an investigation proceeds; or a retraction.
Referral to the institution. Journals do not determine misconduct, and most refer a substantive case to the authors’ institution, which conducts its own inquiry under its own rules and on its own timescale. Authors most often fail to anticipate this step, and it has the widest range of outcomes.
Consequences outside publishing. Subsequent consequences may include funder action, employment consequences and, in clinical disciplines, professional regulatory consequences. Titus, Wells and Rhoades argued in Nature that institutional handling is the decisive variable in research integrity; the practical corollary is that the institution’s process, rather than the journal’s, determines the severity of the outcome.
The escalation sequence, and who controls each step
| Step | Who acts | Possible outcomes | What the author can still influence |
|---|---|---|---|
| 1. Similarity flag | Screening software at submission, or a reader after publication | No further action; or referral to the editor | Everything: the authors can run the check first and disclose overlaps in the cover letter |
| 2. Editorial assessment | The handling editor, reading the matched passages against the sources | Dismissed as incidental; or a letter to the corresponding author | Considerable influence; a specific, candid reply approved by all co-authors can change the outcome at this stage |
| 3. Editorial action | The editor, following COPE Core Practices | Rejection; acceptance after attribution; correction; expression of concern; retraction | Some, by proposing a concrete remedy rather than contesting the score |
| 4. Referral to the institution | The authors’ institution, under its own misconduct procedure | No finding; informal action; formal finding of misconduct; employment consequences | Little; the timescale and the process are not set by the author |
| 5. Funder and regulator | Funding body; in clinical fields, the professional regulator | Loss of funding; publication bans; in one documented UK case, a three-month suspension from medical practice | Almost none |
| 6. The permanent record | Indexing databases | A retraction notice attached to the article, discoverable indefinitely, carried by every co-author | Nothing |
Sequence described from COPE Core Practices and the editorial literature. The sanctions in step 5, including the three-month suspension from medical practice, are those documented by Das N, Panjabi M. Plagiarism: why is it such a big issue for medical writers? Perspectives in Clinical Research. Institutional procedures differ between countries and employers.
Evidence from the retraction record
The best single dataset on outcomes is Fang, Steen and Casadevall’s analysis in PNAS of 2,047 retracted biomedical and life-science articles indexed in PubMed 15. They found 67.4% retracted for misconduct, broken down as 43.4% fraud or suspected fraud, 14.2% duplicate publication and 9.8% plagiarism, against 21.3% for error. They also reported that the share of published articles retracted for fraud or suspected fraud had risen roughly tenfold since 1975, a rise not explained by growth in publication volume alone.
Two conclusions follow. First, plagiarism accounts for a smaller share of the retraction record than its prominence in training suggests, and duplicate publication for a larger one; this is important because duplication is the offence authors are most likely to commit without recognising it 15. Second, a retraction for plagiarism appears in the same category of notice as a retraction for fraud and is indistinguishable from it to a reader who has not examined the breakdown. The distinction that matters to the author is not visible in the record.
The figures cannot establish incidence. Retractions measure detection and editorial willingness to act, so a rising rate is compatible with improving behaviour and better screening 15. The survey evidence has the opposite limitation: Fanelli, in PLoS ONE, pooled self-report data from 21 reviewed studies, 18 of which were meta-analysed, and found 1.97% of scientists admitting to fabricating, falsifying or modifying data and up to 33.7% admitting other questionable research practices; when asked about colleagues, 14.12% reported knowing someone who had fabricated or falsified data and up to 72% reported colleagues’ questionable practices 20. Self-report provides a lower bound; reports about colleagues provide an upper bound inflated by double-counting 20. The true rate is unknown, and quoting either of these figures as a prevalence estimate overstates what it shows.
Retracted biomedical and life-science articles indexed in PubMed as of 3 May 2012. Source: 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.
Documented sanctions
Discussion of consequences often ends at retraction, which is the least severe of them. Das and Panjabi, reviewing the problem for medical writers in Perspectives in Clinical Research, set out the range of sanctions that have been applied in practice: retraction of the affected article, bans on publishing in the journal or with the publisher for a defined period, reports to the author’s employer, loss of research funding and, in one United Kingdom case they describe, a three-month suspension from medical practice 16. The last of these is particularly significant: in clinical disciplines, a publishing offence can affect a professional licence.
Three features of these consequences merit explicit statement because they should inform how authors respond to the first communication from an editor. They are sequential and independent: a journal’s decision to retract does not conclude the institutional process, and an institutional finding of no misconduct does not restore a retracted paper. They are durable: a retraction notice is permanent, indexed and retrievable in any search of the author’s name, unlike a rejection, which leaves no public trace. They are also collective: co-authors who had no part in the copied passages carry the retraction on their own publication records, which is why an honest early explanation to an editor is an obligation to collaborators as well as a prudent course of action.
For junior researchers the career consequence is often the most significant, and it is not primarily formal. A retraction attached to a name during a training period is visible to every subsequent selection committee that searches for it. The Pubrica notes on research integrity cases cover how these processes have run in practice.
Text recycling as a contested practice
Reusing another author’s text without attribution is established misconduct. The reuse of one’s own text is not settled, and claims of a clear consensus in either direction overstate the evidence 8,9.
The most detailed measurement of current practice comes from Anson and Moskovitz, in Accountability in Research, who text-analysed 400 articles from 80 projects funded by the US National Science Foundation, making 800 pairwise comparisons and flagging 1,359 sentence-level instances 6. Every discipline they examined averaged more than three recycled sentences per paper, about 1.2% of a typical 250-sentence paper, while 19.5% of papers showed no detected recycling at all. The distribution was heavily skewed: around 15% of research teams recycled substantially more text than the rest, and one engineering group averaged roughly 50 recycled sentences per paper, about 20% of its articles’ text 6,16. In a sample of 80 paper pairs, only 5 (6.25%) had identical author lists, which is significant because it indicates that most recycling involves text whose other authors are not party to the reuse.
Researchers’ own views are divided. Anson and Moskovitz report that Hall and colleagues found 42% of expert and 56% of novice researchers considered recycling text from prior papers inappropriate, a majority in neither group, with the novices stricter than the experts 6. They also note Citron and Ginsparg’s estimate, from patterns of text reuse in a scientific corpus, that roughly one in eight arXiv papers contained more than 10% recycled content 15. Moskovitz has argued in Science and Engineering Ethics that the practice needs clearer discipline-specific norms rather than blanket prohibition, and COPE’s text recycling guidelines take a similarly graded position: the significance depends on the section, the extent, and whether the reuse is disclosed 2,8. Roig’s guide for the US Office of Research Integrity sets out the conservative end of the range 10.
Given this disagreement, the practicable position is procedural rather than doctrinal. Recycling a few sentences of unchanged procedural description in a methods section, with the earlier paper cited, is widely accepted. Recycling introduction or discussion text is viewed more severely, because these sections contain the intellectual contribution. Recycling any text without citing the earlier work is indefensible regardless of section. Disclosing the overlap to the editor in the cover letter converts the question from a potential problem into an administrative note.
Text-analytic study of recently published research articles from US National Science Foundation–funded projects. Source: Anson IG, Moskovitz C. Text recycling in STEM: a text-analytic study of recently published research articles. Accountability in Research.
Duplicate publication and its consequences
Duplicate publication accounted for 14.2% of the retractions Fang and colleagues analysed, more than plagiarism, and it is the offence authors most often commit while believing that they are doing something else 15. Das and Panjabi note that a PubMed search for the “Duplicate Publication” publication type returned at least 957 records as of February 2011, which is a count of detected cases rather than of occurrences 16.
Editors treat it severely because of the damage done downstream. Von Elm and colleagues, in JAMA, analysed articles used in systematic reviews and identified several distinct patterns of duplicate publication, some of which are extremely difficult for a reviewer to detect because the author list, the framing and the reported outcomes all change between versions 25. Qi and colleagues documented covert duplicate publications in a specific clinical literature in the American Journal of Medicine, and separately showed in PLoS ONE how duplicates propagate across PubMed, EMBASE and the Cochrane Library 13,14. Errami and colleagues built the Déjà vu work on duplicate citations in Medline, reported in Bioinformatics, and Sun and colleagues characterised text similarity across full-text biomedical publications in PLoS ONE 19,22. The cumulative picture is of a literature in which the same participants are counted more than once without disclosure, which inflates pooled effect estimates. Alfonso and colleagues, writing in Revista Española de Cardiología, asked whether the field can afford it; DeAngelis, in JAMA, and Kassirer and Angell and Angell and Relman, in the New England Journal of Medicine, had already set out why editors cannot 23,26,27,28.
The preventive measure is a statement in the cover letter rather than a matter of judgement. Authors should list every related abstract, publication and manuscript under consideration elsewhere, state clearly where a dataset overlaps a previous report, and cite their own earlier work in the text. Researchers conducting a systematic review on the topic rely on authors to do this, and an editor informed in advance rarely objects.
Detection methods and the interpretation of similarity scores
Similarity screening is now routine, and the figure it produces is frequently misunderstood by authors and occasionally by reviewers. Foltynek, Meuschke and Gipp’s systematic literature review in ACM Computing Surveys surveys what academic plagiarism detection systems do: they identify textual and, increasingly, semantic and cross-language overlap against an indexed corpus 7. They cannot determine intent, cannot distinguish a properly quoted and cited passage from an unattributed one without human reading, and cannot see sources outside their index.
In practice, a similarity report must therefore be read rather than scored. A 30% overall score made up of the reference list, a standard instrument description and a correctly quoted definition is unremarkable. A 7% score consisting of two consecutive paragraphs of another author’s discussion is serious. Editors examine the longest matched string, the section in which it occurs, whether the source is cited anywhere, and whether the matched text conveys an interpretation or a procedure. The Pubrica page on similarity screening explains how to read a report before submission, and the types of plagiarism page separates the categories editors distinguish between.
Authors should run the check before submission and act on the matched passages rather than on the total. If a passage is flagged because it is standard procedural wording from the authors’ own earlier paper, that paper should be cited and the overlap disclosed. If it is flagged because it is another author’s sentence, it should be rewritten in original wording and cited, or quoted and cited. If the origin of a passage cannot be recalled, this indicates a weakness in note-taking, and Roig’s guide for the Office of Research Integrity largely concerns the writing habits that prevent it 10.
Inappropriate and appropriate handling of a flagged methods paragraph
Most real cases bear little resemblance to wholesale copying. They resemble the paragraph below: one evaluative sentence absorbed from a review the author read six months earlier, and one procedural passage carried over from the group’s own previous paper because the procedure had genuinely not changed. Neither was intended as plagiarism. Both will be flagged, and only one of them is difficult to defend.
The same methods paragraph, before and after attribution
Two flags, one indefensible
Participants were recruited consecutively from the outpatient clinic. Depression was assessed using the Hamilton Rating Scale for Depression, a 17-item clinician-administered instrument with well-established psychometric properties that has become the gold standard for the assessment of depression severity in clinical trials. Blood samples were drawn after an overnight fast of at least ten hours, centrifuged within thirty minutes and stored at minus eighty degrees Celsius until analysis.
The “gold standard” sentence is someone else’s judgement, reproduced verbatim without attribution. This is the indefensible flag: a borrowed evaluative claim, not a borrowed procedure.
The sampling sentence is word-for-word from the group’s own earlier paper, uncited. Defensible as practice, but the missing citation makes it a problem rather than an administrative note.
Nothing tells the editor an overlap exists, so the overlap is discovered by software after submission rather than declared before it.
A reader cannot tell this is the same cohort as the earlier paper, which is precisely the information a systematic reviewer needs to avoid double-counting participants.
Attributed and disclosed
Participants were recruited consecutively from the outpatient clinic using the procedure described in our earlier report of this cohort (Author et al., 2023). Depression severity was assessed with the 17-item clinician-administered Hamilton Rating Scale for Depression, whose psychometric properties are reviewed by Williams. Blood sampling and storage followed the protocol used in that earlier report: samples were drawn after an overnight fast of at least ten hours, centrifuged within thirty minutes and stored at minus eighty degrees Celsius until analysis. Because the sampling procedure is unchanged, this description reuses wording from Author et al. (2023); the overlap is disclosed in the cover letter.
The earlier paper is cited, which attributes the wording and also alerts a systematic reviewer that the cohorts overlap.
The borrowed assessment is replaced by a citation to the source that makes it.
Reuse of procedural wording is stated openly, which is the graded position COPE’s text recycling guidelines contemplate.
The cover-letter disclosure converts a similarity flag into information the editor has already received.
No methodological detail is lost; the paragraph is longer by one clause.
The science is unchanged. The difference is that both borrowings are now visible: the other author’s judgement is attributed, and the author’s own reuse is cited and declared. That is the entire difference between a paragraph an editor notes and a paragraph an editor investigates. The more serious flag was the evaluative sentence taken from a review, not the procedural passage carried over from the group’s own work, which Anson and Moskovitz found in more than three sentences per paper across every discipline they examined.
Responding to an allegation or to plagiarism identified in published work
Two situations share the same first principle: the response should be prompt and specific.
If an editor writes to the authors. The reply should address the specific passages rather than the overall score. For each flagged passage, it should state the source, how the passage entered the draft, and the proposed remedy: attribution, rewriting or quotation. Authors should not dispute the percentage, should not characterise the overlap as trivial before the editor has done so, and should not send a response that the co-authors have not reviewed. COPE’s Core Practices give editors a graded set of responses, and the choice between them depends substantially on whether the author’s account is credible and complete 4. Where the flag is considered a false positive (a standard instrument description, a correctly quoted passage or a common phrase), the reply should identify the specific text so that the editor can verify it.
If the problem is found in work already published. The authors should write to the journal themselves. A correction initiated by the authors is a different document from a retraction initiated after a reader’s complaint, and editors state this in the notice. The co-authors should be informed first, followed by the institution’s research integrity office; a preprint or repository copy should not be quietly updated, because a divergence between versions is more serious than the original error.
For a co-author with no involvement. The fourth ICMJE authorship criterion commits every author to ensuring that questions about the accuracy or integrity of any part of the work are investigated and resolved; participation in the investigation is therefore an obligation accepted with authorship 1,3. Our page on the ethics of biomedical communication sets out the obligations that this criterion entails.
Effective prevention practices
Prevalence among trainees suggests that this is a writing-skills problem well before it becomes an integrity problem 18,24. Das and Panjabi report a Croatian study in which over 90% of second-year medical students plagiarised to some extent in assigned essays, and a Boston University study finding plagiarism in about 1 of every 20 medical residency application essays — the latter from the work of Segal and colleagues in Annals of Internal Medicine, and the former from Bilić-Žulle and colleagues in the Croatian Medical Journal 6,16. Cicutto, writing in Chest, framed it in the same way: the risk can be avoided through technique rather than exhortation 21.
The relevant techniques are routine and effective. Notes should be taken in original wording from the start, never by pasting; pasted text that enters a draft file is the most common origin of an unattributed sentence. The source, including the page, should be recorded with every note. The methods section should be drafted from the protocol rather than from a previous paper. Where procedural wording must be reused, the earlier work should be cited in the same sentence. Direct quotations should remain visibly marked in the draft until each has been checked. Finally, the similarity check should be run on the manuscript before submission, with sufficient time to act on its findings.
For research teams, two practices matter more than any individual discipline. One person should be responsible for the cover letter’s list of related work, so that overlapping publications are declared rather than assumed to be known. Internal review should also treat statements such as “this paragraph is from our last paper” as routine, because the alternative, a flag at a journal that no team member anticipated, is how a procedural reuse becomes an integrity question. The reporting guidelines for the study design are also a useful drafting prompt, because writing a methods section against a checklist produces original sentences rather than borrowed ones.
Key points
Key points
The consequences of plagiarism occur in sequence and outlast the paper: an editorial assessment, a permanent retraction notice borne by every co-author, a referral to the author’s institution through a process the author does not control, and documented sanctions extending to publication bans, loss of funding and, in one United Kingdom case, a three-month suspension from medical practice. Plagiarism accounts for 9.8% of the 2,047 retractions analysed by Fang and colleagues, whereas duplicate publication accounts for 14.2%; the offence authors are most likely to commit unknowingly therefore has the larger impact. The reuse of an author’s own wording is a contested question without consensus, and the only reliable protection is procedural: citing the earlier work, disclosing the overlap to the editor, and conducting a similarity check with sufficient time to act on the results.
Application to your study
Application to your study
- Conduct a similarity check on the manuscript before submission, allowing sufficient time to address the findings. Examine the matched passages rather than the overall percentage.
- List every related abstract, paper and submission in the cover letter, and state explicitly where a dataset overlaps with a previous report.
- Cite the earlier paper in the same sentence as any reused procedural wording, and add a clause stating that the description is reused.
- Review the introduction and discussion separately and more stringently than the methods, because recycling is judged most severely in these sections.
- Replace any borrowed evaluative sentence with a citation to its source, or quote and attribute it. A judgement about a method is not procedural wording.
- Avoid pasting source text into draft files. Record notes in original wording with the source and page attached; a pasted sentence that loses its provenance is the most common origin of a flag.
- If an editor raises concerns, respond to each passage individually within days, with the co-authors’ review of the reply, and propose a concrete remedy rather than contesting the score.
- If recycled or unattributed text is identified in published work, the authors should contact the journal directly, having first informed the co-authors and the institutional research integrity office.
Frequently asked questions
They occur in sequence and involve several independent bodies. The first is an editorial assessment, which may end in rejection, a correction, an expression of concern or a retraction. For substantive cases, referral to the author’s institution follows, and the institution conducts its own misconduct inquiry on its own timescale. Funder action and, in clinical fields, professional regulatory action may then follow. Das and Panjabi, in Perspectives in Clinical Research, document sanctions that have included retraction, bans on publishing in a journal or with a publisher, reports to employers, loss of funding and a three-month suspension from medical practice in one UK case. The retraction notice itself is permanent, indexed, and carried by every co-author.
Less often than its prominence suggests. Fang, Steen and Casadevall, in PNAS, analysed 2,047 retracted biomedical and life-science articles indexed in PubMed and found plagiarism cited in 9.8%, duplicate publication in 14.2%, and fraud or suspected fraud in 43.4%, with 67.4% retracted for misconduct overall against 21.3% for error. They also found that the share of articles retracted for fraud had risen roughly tenfold since 1975. These figures should be interpreted with caution: retractions measure detection and editorial willingness to act, not the frequency of the behaviour, so they are not prevalence estimates.
There is no consensus on this question. Anson and Moskovitz, in Accountability in Research, analysed 400 articles from 80 funded projects and found every discipline averaging more than three recycled sentences per paper, about 1.2% of a typical 250-sentence paper, with 19.5% of papers showing none. They also report that Hall and colleagues found 42% of expert and 56% of novice researchers considered the practice inappropriate, a majority in neither group. COPE’s text recycling guidelines take a graded position depending on extent, section and disclosure. The defensible course is to cite the earlier paper and disclose the overlap to the editor.
No; the score alone conveys very little to an editor. Foltynek, Meuschke and Gipp’s review in ACM Computing Surveys describes what detection systems do: they find textual overlap against an indexed corpus. They cannot judge intent, cannot distinguish a properly quoted and cited passage from an unattributed one, and cannot see sources outside their index. A 30% score made up of references, a standard instrument description and a correctly cited quotation is unremarkable; a 7% score made up of two consecutive paragraphs of another author’s discussion is serious. Editors examine the longest matched string, the section in which it occurs, and whether the source is cited anywhere.
Because of its effect on anyone pooling the evidence. Von Elm and colleagues, in JAMA, analysed articles used in systematic reviews and identified several patterns of duplicate publication, some very difficult to detect because the author list, framing and reported outcomes change between versions; Qi and colleagues documented covert duplicates in a clinical literature and showed how they propagate across PubMed, EMBASE and the Cochrane Library. As a result, the same participants are counted more than once without disclosure, and pooled effect estimates are inflated. Duplicate publication accounted for 14.2% of the retractions Fang and colleagues analysed, more than plagiarism, and Das and Panjabi note at least 957 PubMed records carrying the “Duplicate Publication” type as of February 2011.
The reply should be prompt and specific, and all co-authors should review it. Each flagged passage should be addressed individually: its origin, how it entered the draft, and the proposed remedy (attribution, rewriting or a marked quotation). Authors should not dispute the overall percentage or describe the overlap as trivial before the editor has characterised it. Where a flag is considered a false positive, such as a standard instrument description or a correctly quoted sentence, the reply should identify the specific text so that the editor can verify it. Under COPE’s Core Practices the editor has a graded set of responses, and the choice between them depends heavily on whether the author’s account is complete and credible.
The authors should contact the journal themselves, before anyone else does. A correction initiated by the authors is a visibly different document from a retraction prompted by a reader’s complaint, and editors state the difference in the notice. The co-authors should be informed first, since the notice will attach to their records as well, and the institution’s research integrity office should be informed without waiting for a request. A preprint or repository copy should not be quietly amended: a divergence between versions is harder to explain than the original error. Under the fourth ICMJE authorship criterion, every author has already agreed to ensure that such questions are investigated and resolved.
Primarily through changes in note-taking, which is where the prevalence evidence points. Das and Panjabi report a Croatian study finding that over 90% of second-year medical students plagiarised to some extent in assigned essays, and a Boston University study finding plagiarism in about 1 of every 20 residency application essays, which suggests a skills problem rather than a problem of character. Notes should be taken in original wording from the start, with the source and page attached; text should never be pasted into a draft file; methods should be drafted from the protocol rather than from a previous paper; the earlier work should be cited in the same sentence as any reused wording; and the similarity check should be run with sufficient time to act on it.
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
- Hall S, Moskovitz C, Pemberton MA. Attitudes toward text recycling in academic writing across disciplines. Accountability in Research. 2018;25(3):142-169. doi:10.1080/08989621.2018.1434622 read 9 October 2026. doi.org/10.1080/08989621.2018.1434622
- Roig M; US Office of Research Integrity. Avoiding plagiarism, self-plagiarism, and other questionable writing practices: a guide to ethical writing. US Office of Research Integrity. 2015. read 9 October 2026. equator-network.org
- 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
- Citron DT, Ginsparg P. Patterns of text reuse in a scientific corpus. Proceedings of the National Academy of Sciences of the USA. 2015;112(1):25-30. doi:10.1073/pnas.1415135111 read 9 October 2026. doi.org/10.1073/pnas.1415135111
- 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 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
- 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
- Segal S, Gelfand BJ, Hurwitz S, Berkowitz L, Ashley SW, Nadel ES, et al. Plagiarism in residency application essays. Annals of Internal Medicine. 2010;153:112-120. 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
- Fanelli D. How many scientists fabricate and falsify research? A systematic review and meta-analysis of survey data. PLoS ONE. 2009;4(5):e5738. doi:10.1371/journal.pone.0005738 read 9 October 2026. doi.org/10.1371/journal.pone.0005738
- Cicutto L. Plagiarism: avoiding the peril in scientific writing. Chest. 2008;133:579-581. read 9 October 2026. pmc.ncbi.nlm.nih.gov
- 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
- 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
- Bilic-Zulle L, Frkovic V, Turk T, Azman J, Petrovecki M. Prevalence of plagiarism among medical students. Croatian Medical Journal. 2005;46:126-131. read 9 October 2026. pmc.ncbi.nlm.nih.gov
- 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
- 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). The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions. Pubrica. https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/
Pubrica Academy. “The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions.” Pubrica, 9 Oct. 2026, pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/.
Pubrica Academy (2026) The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions. Pubrica. Available at: https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/ (Accessed: 15 October 2026).
Pubrica Academy. “The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions.” Pubrica. October 9, 2026. https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/
Pubrica Academy, ‘The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions’ (Pubrica, 9 October 2026) <https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/> accessed 15 October 2026.
Pubrica Academy. The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions. [Internet]. Pubrica; 2026 Oct 9 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/
Pubrica Academy, “The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions,” Pubrica, Oct. 9, 2026. [Online]. Available: https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/
@misc{what_are_the_consequences_of_plagiaris_2026,
author = {{Pubrica Academy}},
title = {The Consequences of Plagiarism in Scientific Writing: Investigation, Retraction and Sanctions},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/medical-writing/what-are-the-consequences-of-plagiarism-in-scientific-writing/},
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.
Pre-submission originality and overlap declaration
This record should be completed before submission and kept with the manuscript. Blocks A and B are for the authors’ own file; block C is the text to be adapted into the cover letter, where declared overlap ceases to be a problem. The most valuable line on the page is the list of related work.
ORIGINALITY AND OVERLAP DECLARATION
Manuscript: ______________________ Target journal: ______________________
Completed by: ______________________ Date: __________
A. SIMILARITY REPORT REVIEW (read the passages, not the score)
Overall similarity: ____% (this number on its own means nothing)
Longest single matched passage: ____ words, in section: ____________
For EACH match over about 25 words:
Passage: ______________________________________________
Source: ______________________________________________
Is the source cited in the manuscript? YES / NO
Category: [ ] our own earlier work (procedural)
[ ] our own earlier work (introduction / discussion)
[ ] another author's procedural wording
[ ] another author's evaluative or interpretive sentence
[ ] reference list / standard instrument name / common phrase
Action: [ ] cite in the same sentence [ ] rewrite [ ] quote and cite
[ ] no action needed, and why: __________________________
B. SECTION-BY-SECTION CHECK (stricter as you go up this list)
[ ] Discussion contains no reused wording from any earlier paper
[ ] Introduction contains no reused wording from any earlier paper
[ ] Results prose is written fresh for this dataset
[ ] Methods reuse, where present, cites the earlier paper in the same sentence
[ ] No borrowed evaluative claims ("gold standard", "well established",
"the most widely used") without a citation to whoever made them
[ ] Every direct quotation is marked and attributed
[ ] Every reference has been checked against the original source, not a
citation manager export
C. COVER LETTER TEXT (adapt and include -- do not omit)
Related work by these authors:
Conference abstracts presenting these data: ______________________
Published papers using this dataset or cohort: ______________________
Manuscripts under consideration elsewhere: ______________________
Preprints deposited: ______________________
Dataset overlap: the participants reported here [do / do not] overlap with
those in ______________________. The extent of overlap is ____________.
Text overlap: the methods section reuses procedural wording from
______________________, cited in the text, because the procedure is
unchanged. No other section reuses previously published wording.
[ ] All co-authors have seen and approved this declaration
D. IF AN ALLEGATION ARRIVES
[ ] Do not reply before co-authors have seen the draft response
[ ] Address each flagged passage individually: origin, route into the draft,
proposed remedy
[ ] Do not contest the percentage; propose a concrete fix
[ ] Notify the institutional research integrity office the same week
[ ] Do not amend any preprint or repository copy while the matter is open 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 editing team reviews similarity reports as a handling editor would: the longest matched string first, then the section in which it occurs, then whether the source is cited. Pubrica distinguishes administrative flags from substantive ones, revises the passages that require revision, drafts the overlap declaration for the cover letter, and checks self-citation so that a systematic reviewer can identify where datasets overlap. Where an allegation has already been made, Pubrica assists in drafting a response that addresses the passages rather than the score.