A retraction does not just remove a paper. It removes trust in the authors, the sponsor, and every study that cited the work. With retraction numbers rising across biomedical publishing, publication ethics has moved from a compliance footnote to a core competency for anyone producing scientific manuscripts.
Why retractions are rising
Journals retracted more papers in the last few years than in any comparable period before, and the growth is not only because more misconduct is happening. Detection has improved. Image forensics software flags duplicated western blots, plagiarism screening catches recycled text across languages, and post-publication platforms give readers a public venue to raise concerns. At the same time, genuine pressure to publish has fed an industry of paper mills selling authorship on fabricated manuscripts, and some of that output reaches indexed journals before it is caught.
For pharmaceutical, biotech, and device companies, the stakes are specific: a retracted publication can undermine a regulatory submission that cited it, weaken a health technology assessment dossier, and hand competitors and litigators a documented credibility problem.
The most common reasons papers are retracted
- Data problems. Fabrication, falsification, and image manipulation remain the leading causes of misconduct retractions, but honest errors in analysis or transcription also force retractions when they change conclusions.
- Plagiarism and duplicate publication. Reusing text without attribution, and submitting substantially the same study to two journals, both violate journal policies even when the underlying data are sound.
- Authorship disputes and ghost authorship. Adding authors who did not contribute, omitting people who did, or concealing professional writing support all breach International Committee of Medical Journal Editors (ICMJE) requirements.
- Undisclosed conflicts of interest. Funding sources and financial relationships that surface after publication are a frequent trigger for corrections and, when material, retractions.
- Compromised peer review. Fake reviewer email addresses and paper mill submissions have caused mass retractions at several publishers.
What ICMJE authorship actually requires
ICMJE sets four criteria, and an author must meet all of them: substantial contribution to the conception, design, data acquisition, or analysis of the work; drafting or critically revising it for important intellectual content; approval of the final version; and agreement to be accountable for all aspects of the work. Contributors who fall short of all four, including professional medical writers, belong in the acknowledgements with their funding source named, not on the author byline.
The practical discipline is to settle authorship in writing before the first draft exists. Most authorship disputes we see arise because roles were never documented and expectations diverged silently over months of drafting.
GPP 2022 and professional medical writing support
Good Publication Practice (GPP 2022) is the current standard for company sponsored publications, and it is explicit on a point that is still widely misunderstood: professional medical writing support is legitimate and beneficial when it is disclosed. Ghostwriting means concealment. Transparent writing support means the writer is named, the funder is named, the authors direct the content and take accountability, and the writer works to the data rather than to a marketing message. A disclosed medical writer working under GPP 2022 typically improves reporting quality, because trained writers follow CONSORT, PRISMA, and journal requirements more consistently than ad hoc drafting does.
Safeguards that actually prevent retractions
- Verify against source data. Every number in the manuscript should trace to the statistical output or clinical study report, and someone other than the drafter should perform that check.
- Screen before submission. Run plagiarism screening on the final manuscript, including your own team's earlier publications, since self plagiarism triggers the same journal policies.
- Document authorship and contributions. Use a publication agreement recording who qualifies as an author, who is acknowledged, and who approved each draft.
- Disclose completely. Funding, writing support, and conflicts of interest are cheap to disclose and expensive to have discovered.
- Choose journals with care. Predatory journals with no real peer review damage credibility even without a retraction. Check indexing, editorial boards, and Committee on Publication Ethics (COPE) membership before submitting.
- Register and report consistently. Outcomes reported in the manuscript should match the trial registration; unexplained outcome switching is one of the fastest ways to attract post-publication scrutiny.
When something goes wrong anyway
Errors discovered after publication are not automatically retractions. COPE guidance distinguishes corrections, which fix errors that do not alter conclusions, from expressions of concern and retractions, which apply when findings are unreliable. The worst response is silence. Authors and sponsors who approach the journal promptly, with a clear account of the error and its impact, consistently fare better than those who wait for readers to find it.
Predatory journals: the credibility trap that needs no misconduct
A manuscript can be honest, accurate, and still damaged by where it appears. Predatory journals solicit aggressively, promise implausibly fast peer review, and publish for a fee with little or no editorial scrutiny. Publications in such venues rarely count for regulatory or HTA purposes, and their presence on an author's record raises questions about everything else on it. Practical screening takes minutes: check whether the journal is indexed in MEDLINE, Scopus, or the Directory of Open Access Journals, whether the editorial board members are real and aware of their role, whether the publisher belongs to COPE, and whether published articles in the journal show evidence of actual review. Unsolicited invitations praising your earlier work in generic terms are the most reliable warning sign in the inbox.
Registration, reporting, and data availability
Publication ethics now extends past the manuscript into the infrastructure around it. Trials should be registered prospectively, and the outcomes reported in the paper must match the registered outcomes, with any changes explained rather than hidden. Reporting guidelines are part of the same discipline: CONSORT for randomized trials, STROBE for observational studies, PRISMA for systematic reviews, and CHEERS for economic evaluations, each with a checklist journals increasingly require at submission. Data availability statements are becoming standard, and sponsors should decide their data sharing position before submission rather than improvising it in response to a reviewer. A paper whose registration, reporting checklist, and data statement all align gives post-publication critics very little to work with.
AI tools and authorship disclosure
Journal policies have converged quickly on two rules: artificial intelligence tools cannot be authors, because they cannot take accountability, and their use in drafting must be disclosed in the methods or acknowledgements. The ICMJE recommendations now address AI assisted writing directly, and several major publishers require a specific statement. The safe operating position for sponsored publications is straightforward. Document where AI assistance was used, verify every claim and every reference by human review against source material, and disclose per the target journal's policy. Concealment converts an ordinary tool into a misconduct allegation waiting to happen.
How EvySaif approaches publication integrity
Every manuscript we support is written by a clinician led team to GPP 2022 and ICMJE standards, with documented authorship agreements, source data verification, plagiarism screening, and full disclosure of our role. That rigor is not overhead. It is what keeps a publication usable in the submissions, dossiers, and scientific discussions it was written to support. Learn more about our scientific and medical writing services or the standards behind our evidence synthesis work.