Ethical declarations for journal submission

Key Takeaways:

Most journals now expect 8 to 12 separate ethical declarations, not just an ethics approval statement; a single missing declaration is one of the most common causes of desk rejection.
The January 2026 ICMJE revision added a dedicated section on artificial intelligence, so AI use disclosure now sits alongside competing interests and funding as a standard submission requirement.
The same declaration often has to be entered in 3 places: the cover letter, the manuscript end matter, and the submission portal; mismatches between them trigger integrity queries.
Build a reusable declarations pack using the templates, publisher comparison tables, and pre-submission checklist in this guide, then adapt it for each journal.

Table of Contents

Glossary of Key Terms

Term	What it means
Ethical declaration	A formal statement confirming how a study was approved, funded, conducted, and reported, submitted alongside the manuscript.
IRB / REC	Institutional Review Board or Research Ethics Committee: the body that reviews and approves research involving humans.
IACUC / AWERB / CPCSEA	Committees that review and approve research involving animals in the United States, the United Kingdom, and India respectively.
Informed consent	Voluntary agreement to take part in a study, given after the participant understands its purpose, risks, and benefits.
Consent for publication	Separate permission to publish identifiable material such as photographs, video, case details, or genomic data.
ICMJE	International Committee of Medical Journal Editors, which issues the most widely adopted submission and authorship rules.
COPE	Committee on Publication Ethics, which publishes flowcharts and guidance editors use to handle suspected misconduct.
WAME and CSE	World Association of Medical Editors and Council of Science Editors: 2 further bodies that set editorial policy standards.
CRediT	Contributor Roles Taxonomy: a standard list of 14 roles used to describe exactly what each author contributed.
Ghost authorship	Omitting someone who meets authorship criteria, often a professional writer or a sponsor employee.
Guest or gift authorship	Listing someone who did not meet authorship criteria, often a senior colleague or department head.
Competing interest	Any financial, personal, professional, or ideological relationship that could bias the work or its evaluation.
Data availability statement	A statement telling readers whether the underlying data exist, where they are held, and on what terms they can be accessed.
FAIR principles	A standard for making data Findable, Accessible, Interoperable, and Reusable.
Salami slicing	Splitting 1 study into several thin papers to inflate a publication list.
Duplicate publication	Republishing substantially the same work without disclosure and permission.
Secondary publication	Permitted republication, usually in another language, under conditions set by ICMJE.
Preprint	A manuscript posted publicly on a server such as medRxiv, bioRxiv, arXiv, or SSRN before peer review.
Preregistration	Depositing a study plan and analysis strategy publicly before data collection begins.
Registered Report	A publication format in which the methods are peer reviewed and accepted in principle before results exist.
EQUATOR Network	A registry of reporting guidelines and checklists covering most study designs.
ARRIVE 2.0	The reporting guideline for research involving live animals.
3Rs	Replacement, Reduction, and Refinement: the framework for minimizing animal use and suffering.
SAGER guidelines	Sex and Gender Equity in Research: guidance on reporting sex and gender in study design and analysis.
CARE Principles	Collective benefit, Authority to control, Responsibility, and Ethics: governance principles for Indigenous data.
Helicopter research	Research conducted in a country or community without meaningful local partnership, credit, or benefit.
DURC	Dual-Use Research of Concern: research that could be misused to cause serious harm.
De-identification	Removing or masking information that could identify an individual in data or images.
Expression of concern	A public editorial notice flagging unresolved doubts about a published article.
Retraction	Permanent withdrawal of a published article from the scholarly record, with a public notice explaining why.
Paper mill	An organization that sells fabricated manuscripts, data, or authorship positions.
Trial registration number	The unique identifier issued when a clinical trial is registered on ClinicalTrials.gov, CTRI, ISRCTN, or another WHO ICTRP primary registry.

What are ethical declarations, and why do journals reject papers over them?

Ethical declarations are formal statements confirming how your study was approved, funded, conducted, and reported. Journals reject papers when a declaration is missing, vague, or contradicts something else in the submission.

Editors treat declarations as evidence of accountability rather than paperwork. A statement that cannot be verified is worse than no statement at all, because it raises questions about everything else in the manuscript. Most declaration-related rejections happen before peer review, at the editorial screening stage, where staff check the submission against a fixed compliance list.

The most frequent reasons manuscripts are returned or desk rejected are listed below.

The ethics approval statement names no committee, no approval number, and no date.
The paper reports human participants but declares no consent process.
Identifiable photographs or case details appear without consent for publication.
The competing interests section is blank rather than saying there are none.
Funding appears in the acknowledgments but not in a funding statement, or grant numbers are missing.
A clinical trial was registered after enrollment began, or the registration number is absent.
Use of AI tools is visible in the writing but is not disclosed anywhere.
The data availability statement says data are available on request, with no contact point or conditions.
A preprint exists but was not declared, so screening software flags the text as duplicated.
Author contributions do not match the author list, or a listed author cannot be reached.

The Core Ethical Declarations Every Author Must Submit

The declarations below apply across most journals and disciplines. Later sections cover discipline-specific additions, publisher differences, and ready-to-use templates.

1. Ethics approval and institutional review statements

This declaration confirms that an independent committee reviewed and approved your protocol before the work began. It applies to research involving people, human tissue, identifiable personal data, and animals.

Include the following details.

Full name of the approving committee, IRB, or ethics board, including the institution it belongs to.
The approval reference or protocol number exactly as issued.
The date of approval, and confirmation that it preceded data collection.
The name of the framework followed, such as the Declaration of Helsinki for human research.
For multi-site studies, approval details for every site, or the name of the lead committee under a single-review arrangement.
For exempt or waived studies, the committee that granted the exemption and the ground on which it was granted.

2. Informed consent to participate

Consent to participate confirms that each participant freely agreed to take part after being told what the study involved. It is a separate requirement from ethics approval, and journals expect both.

State whether consent was written, verbal, or electronic, and why, if it was not written.
For participants under 18 years of age, state that consent came from a parent, guardian, or legal representative, and record assent from the child where appropriate.
For participants who cannot consent for themselves, name the process used with their legally authorized representative.
Where a committee waived consent, for example in a retrospective record review, state the waiver and the committee that granted it.
Describe consent for secondary use of stored data or biological samples separately, because reuse consent is now scrutinized closely.

3. How is consent for publication different from consent to participate?

Consent to participate covers taking part in the study. Consent for publication covers making identifiable material public: photographs, video, audio, case histories, pedigrees, or genomic data. Most journals require both.

Removing a name is not enough. A case report can identify someone through a rare diagnosis, a date, a hospital, or an occupation. Cropping eyes from a facial photograph is no longer accepted as anonymization by most publishers.

Obtain publication consent on the journal or publisher form where one exists.
Show the participant the version of the material that will be published, not a summary of it.
Retain the signed form; do not send it with the submission unless asked, because it contains personal data.
If the participant has died or cannot be traced, contact the journal before submitting; a next-of-kin process may apply.
For images of children, obtain consent from a parent or guardian and note this explicitly.

4. Authorship and contributorship declarations

Authorship disputes are among the most common ethics cases editors handle. A clear contributorship declaration at submission prevents most of them.

The ICMJE sets 4 criteria, and an author should meet all 4.

Substantial contribution to conception or design, or to acquisition, analysis, or interpretation of data.
Drafting the work or revising it critically for important intellectual content.
Final approval of the version to be published.
Agreement to be accountable for all aspects of the work, including the accuracy and integrity of any part of it.

Contributions that do not meet all 4 criteria belong in the acknowledgments, not the author list. Common examples include funding acquisition alone, general supervision of the research group, or technical assistance.

The CRediT taxonomy gives you a standard vocabulary of 14 roles for describing who did what, and many journals now require it.

Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation.
Writing (original draft), Writing (review and editing), Visualization, Supervision, Project administration, Funding acquisition.

Watch for the 3 authorship problems editors screen for most often.

Ghost authorship: someone who meets the criteria, often a medical writer or sponsor employee, is left off the list.
Guest or gift authorship: a senior colleague, department head, or collaborator is added without qualifying contributions.
Authorship for sale: positions bought from a paper mill, which is treated as serious misconduct and usually ends in retraction.

Also declare the practical points that journals now check.

The corresponding author, who handles communication, proofs, and post-publication queries.
Equal contribution statements, marking which authors share first or last position.
ORCID identifiers for every author, which several publishers require for the corresponding author at minimum.
Group or consortium authorship arrangements, including who is searchable in PubMed and who is listed in a collaborator group.
Any change to the author list after submission, which needs written agreement from every author, including the person being added or removed.

5. Declaration of competing interests

A competing interest is any relationship that a reasonable reader might think could have influenced the work. Having one is not misconduct; failing to disclose one is.

Disclose interests in these categories.

Financial: employment, consultancies, honoraria, stock or share options, patents, expert testimony, paid travel, and speaker fees.
Funding: grants, contract research, equipment or drug donations, and in-kind support from a party with an interest in the results.
Personal: family relationships, close friendships, and rivalries with people involved in the work or its review.
Professional: unpaid advisory roles, board memberships, membership of advocacy organizations, and editorial positions at the journal.
Intellectual: strongly held published positions that the paper defends or attacks.

Cover the standard 36-month look-back period used by the ICMJE disclosure form, and include interests held by immediate family members where the form asks for them. If there are none, say so explicitly; a blank section is treated as an omission.

6. Funding and financial support statements

A funding statement is not the same as a competing interest statement, and journals want both. The funding statement records who paid for the work; the competing interest statement records what that relationship might mean for objectivity.

Name every funder in full, using the registered name from the Crossref Funder Registry so the record can be linked automatically.
Give the exact award or grant number for each funder.
Add institutional identifiers such as ROR IDs where the journal requests them.
State the role of the funder in study design, data collection, analysis, interpretation, and the decision to publish, or state that the funder had no such role.
Declare internal or departmental support and scholarships, which authors often forget.
State explicitly if the work received no specific funding.

Check the mandates attached to your funding before you choose a journal, because they affect licensing and deposit obligations.

Funder or policy	Typical requirement	What to declare
NIH	Deposit the accepted manuscript in PubMed Central; public access on acceptance under current policy.	Grant number, PMCID once issued, and the deposit route used.
Wellcome	Immediate open access under a CC BY license; preprint expected for some calls.	Grant number, license chosen, and preprint DOI if posted.
Plan S and cOAlition S funders	Immediate open access with no embargo; rights retention language in the submission.	Funder name, grant number, and the rights retention statement.
European Research Council	Open access to publications and, where possible, to underlying data.	Project acronym, grant agreement number, and data repository details.
Industry sponsors	Author access to the full dataset and freedom to publish are increasingly required by editors.	Sponsor role, data access arrangements, and any publication approval clause.

7. Do you have to disclose AI use in your manuscript?

Yes. Journals following the January 2026 ICMJE Recommendations require disclosure of AI and AI-assisted tools used in drafting, editing, translation, image generation, or analysis, in both the cover letter and the manuscript.

The 2026 revision added a dedicated section on AI in publishing that applies to authors, reviewers, and editors. The principles below are now close to universal across major publishers, although the level of detail requested varies.

AI tools cannot be listed as an author or co-author, because they cannot take responsibility for accuracy, integrity, or originality.
Human authors remain fully accountable for every part of the manuscript, including anything an AI tool produced.
AI-generated material must not be cited as a primary source, and every factual claim and reference must be verified
Failure to disclose AI use may be treated as misconduct rather than an oversight.
Peer reviewers must not upload manuscripts to AI tools that cannot guarantee confidentiality.
AI-generated or AI-altered images are prohibited by most journals except where the image is itself the object of study.

Write the disclosure so a reader can tell what the tool did and what you did. Name the tool, the version, the date range of use, and the specific task. Language polishing, reference formatting, and code generation are all disclosable uses.

If your study is about AI or machine learning rather than assisted by it, use the reporting checklists built for that purpose, including TRIPOD+AI for prediction models and CLAIM for medical imaging.

8. Data availability statements

A data availability statement tells readers whether the data behind your results exist, where they are held, and on what terms they can be accessed. Most major publishers now require one in every research article, including those with no shareable data.

Aim for the FAIR principles: data that are Findable, Accessible, Interoperable, and Reusable. In practice that means a persistent identifier, a named repository, a license, and enough metadata for someone else to interpret the files.

Situation	What to write	Common mistake
Data are in a public repository	Name the repository, the accession or DOI, and the license.	Giving a personal website or a link that expires.
Data are in the article or supplement	State that all data generated or analyzed are included in the published article and its supplementary files.	Claiming this when only summary tables are included.
Data are sensitive or identifiable	Explain the restriction, name the committee or law that imposes it, and describe the controlled access process.	Saying only that data are confidential, with no route to access.
Data are available on request	Name the person or office, give an institutional contact route, and state the conditions and expected response time.	Writing the phrase alone, which editors increasingly reject.
Third-party data were used	Name the source, the license or agreement, and how another researcher could obtain the same data.	Implying you can share data you do not own.
No new data were generated	State this clearly and point to the sources analyzed.	Leaving the statement blank.

9. Code, software, and materials availability

Where results depend on analysis code, simulations, or custom software, journals increasingly ask for a separate availability statement. Reviewers may request the code during review.

Deposit code in a versioned repository and archive a release with a DOI so the exact version used is citable.
State the license, and choose one that permits reuse if your funder requires it.
List the software, packages, and version numbers needed to reproduce the analysis.
For biological materials, cell lines, plasmids, or reagents, state how other researchers can obtain them and note any material transfer agreement.
Authenticate and declare the provenance of cell lines, and check them against registries of misidentified lines.

10. Clinical trial registration statements

Prospective registration is a condition of publication at ICMJE-following journals, not a formality completed later. The January 2026 revision tightened the expectations around registration timing.

Register before the first participant is enrolled, in a WHO ICTRP primary registry or on ClinicalTrials.gov.
Common registries include ClinicalTrials.gov, CTRI in India, ISRCTN, ANZCTR, ChiCTR, and the EU Clinical Trials Information System.
Place the registration number and registry name at the end of the abstract, where most journals require it.
Report any change to the primary outcome, and explain when and why it changed.
Include a data-sharing plan, which the ICMJE requires for clinical trials as part of registration and reporting.
Follow CONSORT for the report itself and SPIRIT for the protocol.
For systematic reviews, register on PROSPERO and give the registration number.

Retrospective registration is not always fatal, but it must be disclosed openly, with the registration date and an explanation. Concealing it is treated far more seriously than the lapse itself.

11. Prior publication, duplicate submission, and salami slicing

Journals need to know everything about the manuscript history before they invest in peer review. Disclose at submission rather than waiting to be asked.

Confirm the manuscript is not under consideration anywhere else; simultaneous submission to 2 or more journals is prohibited.
Declare any preprint, including the server, DOI, and posting date.
Declare conference abstracts, posters, and proceedings papers, and state how much overlap exists.
Declare thesis or dissertation chapters that the manuscript draws on.
Declare related papers from the same dataset that are published, in press, or under review, and attach them if asked.
Avoid salami slicing: splitting 1 coherent study into several thin papers to lengthen a publication list.
For secondary publication, usually translation into another language, follow the ICMJE conditions, which include permission from both editors and a footnote citing the original.

12. Copyright, licensing, and permissions declarations

Reusing published material without written permission is a copyright breach, and it is caught routinely by editorial screening.

Obtain written permission for every figure, table, image, questionnaire, scale, or long quotation reproduced from another source.
Use the rights clearance service of the original publisher and keep the license number.
Add the required credit line to the caption in the exact wording the license specifies.
Check whether adapting a figure still needs permission; usually it does.
Confirm that your chosen license, such as CC BY or CC BY-NC, is compatible with your funder mandate.
Check your institution or employer does not hold rights that limit what you can sign.
Confirm the self-archiving terms, including which version you may deposit and after what period.

13. Acknowledgments and third-party contributions

Acknowledgments are an ethical declaration in their own right, because they record contributions that did not qualify for authorship.

Obtain written permission from everyone you name, since inclusion can imply endorsement of the findings.
Name medical writers and editorial assistance providers, state who paid for them, and follow Good Publication Practice where the study is industry-sponsored.
Acknowledge statisticians, technicians, data managers, and language editors whose input fell short of the authorship criteria.
Acknowledge participating communities, patient advisers, and data custodians where relevant.
Do not use the acknowledgments to hide a funder or a competing interest that belongs in its own statement.

Which ethical frameworks and bodies govern these declarations?

The Declaration of Helsinki, the Belmont Report, the CIOMS guidelines, and the Nuremberg Code set the underlying principles. ICMJE, COPE, WAME, and CSE translate those principles into the specific statements journals ask for.

Framework or body	What it covers	Why it matters at submission
Nuremberg Code (1947)	The founding statement that voluntary consent is essential in human experimentation.	The origin of the consent requirement journals still enforce.
Declaration of Helsinki (WMA, revised 2024)	Ethical principles for medical research involving human participants.	Most journals require you to state that the study complied with it.
Belmont Report (1979)	Respect for persons, beneficence, and justice, the basis of United States IRB review.	Underpins the exempt and expedited review categories you may cite.
CIOMS guidelines	International ethical guidelines for health-related research, with a focus on low-resource settings.	Used to judge cross-border and multi-country studies.
WMA Declaration of Taipei	Ethics of health databases, biobanks, and reuse of data and samples.	Relevant to secondary data analysis and biobank studies.
ICMJE Recommendations (revised January 2026)	Authorship, disclosure, trial registration, data sharing, AI use, and editorial process.	The single most widely adopted rulebook for submissions.
COPE	Flowcharts and guidance for handling suspected misconduct.	Determines what happens if a declaration is later found to be wrong.
WAME and CSE	Editorial policy statements, predatory publishing guidance, and integrity standards.	Shapes journal-level policies you must comply with.
EQUATOR Network	A registry of reporting guidelines for every major study design.	Journals often require the matching checklist as a submission file.

What changed in the 2024 Declaration of Helsinki?

The 2024 revision replaced the word subjects with participants, extended responsibility beyond physicians to everyone involved in research, and rewrote the rules on consent for stored data and biological materials.

Terminology was updated throughout, and gender-neutral language was adopted.
The Declaration now addresses all individuals, teams, and organizations involved in medical research, not physicians alone.
Scientific integrity was written in for the first time, with a clear position against research misconduct.
Vulnerability is treated as contextual and dynamic rather than a fixed property of certain groups, with an emphasis on fair inclusion.
New provisions call for meaningful engagement with participants and their communities before, during, and after a study.
Paragraph 32 was rewritten to cover consent for collection, processing, storage, and secondary use of biological materials and data.
The risk of re-identifying data that were previously de-identified is flagged explicitly.
Cross-border studies attract dual ethical review expectations.
The principles continue to apply during public health emergencies.

If your manuscript still cites the 2013 version, update it. Editors notice, and the change signals that your consent and data reuse processes may also be out of date.

What changed in the January 2026 ICMJE Recommendations?

The January 2026 revision added a new section on AI in publishing, tightened clinical trial registration timing, strengthened author access to data in sponsored studies, and updated guidance on funding, affiliations, and retractions.

A dedicated AI section, Section V, covering authors, peer reviewers, and editors.
Confirmation that AI tools cannot be authors, and that human authors carry full responsibility for AI-assisted content.
Explicit disclosure requirements for AI use in drafting, editing, translation, image generation, and analysis.
Stronger requirements for author access to the full dataset in industry-funded and collaborative research.
Greater emphasis on sharing de-identified data so results can be verified independently.
Tighter expectations on when a clinical trial must be registered.
Updated rules on listing funding sources and author affiliations.
Revised signposting to guidance on retractions and misconduct handling.

Which reporting guideline applies to your study design?

Find the matching checklist through the EQUATOR Network: CONSORT for trials, PRISMA for systematic reviews, STROBE for observational studies, ARRIVE 2.0 for animal work, and CARE for case reports.

A completed reporting checklist is now a routine submission file. It is not an ethical declaration in the narrow sense, but journals treat an incomplete checklist as a transparency failure, and many titles will not send a manuscript out for review without one.

Guideline	Study type it covers
CONSORT	Randomized controlled trials, with extensions for cluster, pilot, and non-inferiority designs.
SPIRIT	Clinical trial protocols.
PRISMA	Systematic reviews and meta-analyses, with extensions for scoping reviews and network meta-analyses.
MOOSE	Meta-analyses of observational studies.
STROBE	Observational studies: cohort, case-control, and cross-sectional.
STARD	Diagnostic accuracy studies.
TRIPOD+AI	Prediction model studies, including those using machine learning.
CLAIM	Artificial intelligence applied to medical imaging.
CARE	Case reports.
ARRIVE 2.0	Research involving live animals.
COREQ and SRQR	Qualitative research, including interview and focus group studies.
CHEERS	Health economic evaluations.
SQUIRE	Quality improvement studies.
SAGER	Reporting of sex and gender across all designs.
CHERRIES	Web and internet-based surveys.

What declarations does animal research need?

Animal studies have their own approval, husbandry, and reporting declarations. Naming the committee is only the starting point.

Name the animal ethics committee that approved the work, with the protocol number and approval date. Examples include IACUC in the United States, AWERB in the United Kingdom, and CPCSEA in India.
Cite the national legislation followed, such as the Animals (Scientific Procedures) Act or the applicable European Union directive.
State compliance with ARRIVE 2.0 and submit the completed checklist.
Describe how the 3Rs were applied: Replacement, Reduction, and Refinement.
Report species, strain, sex, age, weight, source, and health status.
Describe housing, husbandry, enrichment, acclimatization, and the light cycle.
State anesthesia, analgesia, and euthanasia methods, with doses and routes.
Define humane endpoints and report how many animals reached them.
Give sample size justification, randomization method, and blinding of outcome assessment.
Report all animals excluded from analysis and the reason for each exclusion.

Data Protection and Privacy Declarations

Ethics approval does not discharge your legal obligations under data protection law. Journals increasingly ask for both.

Regime	Applies to	What to declare
GDPR (European Union and United Kingdom)	Personal data of people in the EU and UK, including pseudonymized data.	Lawful basis for processing, data controller, and whether a data protection impact assessment was done.
HIPAA (United States)	Protected health information held by covered entities.	De-identification method used: Safe Harbor removal of the 18 identifiers, or expert determination.
DPDP Act (India)	Digital personal data of individuals in India.	Consent notice used, purpose limitation, and any exemption relied on for research.
Institutional data agreements	Data obtained from registries, hospitals, or third parties.	The agreement in place, permitted uses, and publication restrictions.

De-identification is a claim you must be able to defend. Small cell sizes, rare conditions, precise dates, and geographic detail can all re-identify a person when combined, and the 2024 Declaration of Helsinki specifically flags this risk. State how you assessed it.

Preprints, Preregistration, and Registered Reports

These 3 practices are separate, and each carries its own disclosure obligation.

Practice	What it is	What to declare
Preprint	A manuscript posted publicly before peer review on a server such as medRxiv, bioRxiv, arXiv, SSRN, or Research Square.	Server name, DOI, posting date, and the license applied; check the journal accepts preprinted submissions.
Preregistration	A public, timestamped record of hypotheses, design, and analysis plan, filed before data collection.	Registry name, registration DOI or number, and any deviation from the registered plan.
Registered Report	A format in which methods are peer reviewed and accepted in principle before results exist.	Stage 1 acceptance details and confirmation that the approved protocol was followed.

Deviations from a preregistered plan are acceptable if you report them; concealing them is not.
Label exploratory analyses as exploratory, and keep them separate from confirmatory tests.
Check whether updating a preprint after acceptance is permitted or required by your journal.
Some funders now expect preprint posting; check before you choose a server or license.

Equity and Inclusion Declarations

A growing number of journals require a statement on how the research team was constituted and how affected communities were involved. These declarations are relatively new, so many authors miss them entirely.

Global research and helicopter research statements

If your study was carried out in a country other than your own, expect to explain the local partnership behind it.

State how local researchers were involved in design, implementation, analysis, and authorship.
Confirm that local ethics review took place in addition to any review at your home institution.
Explain how the research is relevant to local priorities and how findings will be returned to the community.
Note capacity-building elements such as training, equipment, or shared infrastructure.
Cite local and regional literature rather than only work from high-income countries.

Indigenous data governance and the CARE Principles

Where research involves Indigenous peoples or their data, FAIR principles alone are not sufficient. The CARE Principles add a governance layer.

Collective benefit: the research and its data should benefit the community that provided them.
Authority to control: communities decide how their data are used, shared, and represented.
Responsibility: researchers must show how the relationship supports self-determination.
Ethics: the rights and wellbeing of the community are the primary concern throughout the data lifecycle.
Declare community consent, data sovereignty arrangements, and any oversight body involved.

Sex, gender, race, and ethnicity reporting

Follow the SAGER guidelines: report sex and gender in the design, analysis, and reporting, and use the 2 terms accurately.
State whether findings apply to all participants or only to a subgroup.
Define race and ethnicity categories, explain who assigned them, and justify their use as variables.
Report the age range and any age-based exclusions, and justify excluding pregnant participants or older adults.
Explain how the study population reflects, or fails to reflect, the population affected by the condition.

How do publishers check image integrity and dual-use risk?

Publishers now screen images automatically before peer review. Declaring your image handling process protects you from an honest error being read as manipulation.

Retain unprocessed original files for every figure, and be ready to supply them on request.
Apply adjustments such as brightness and contrast to the whole image, never to a selected region.
Disclose any splicing in a blot or gel with clear dividing lines and a note in the legend.
Do not duplicate a panel across figures or papers without explicit labeling.
State the microscope, camera, magnification, and processing software used.
Confirm that no AI tool generated or altered a scientific image.

Dual-use research of concern needs its own risk statement. Studies in areas such as pathogen enhancement, toxin synthesis, or offensive security research may require institutional biosafety committee review, and some journals run an additional editorial risk assessment before publication.

Declare biosafety level, containment, and institutional biosafety committee approval.
State any export control, pathogen access, or material transfer restrictions.
Explain what methodological detail has been withheld and why, if anything has been.
For security research, describe the responsible disclosure process followed with the affected vendor.

What if you did not get ethics approval before you started?

Contact your institutional committee immediately and ask for a retrospective review, a formal waiver, or written confirmation that the study was exempt. Do not write an approval statement you cannot document.

Committees can often issue a determination after the fact for low-risk work, particularly retrospective record reviews, service evaluations, and analyses of fully anonymized existing datasets. What editors will not accept is silence or a vague claim that approval was not needed.

Ask the committee for a written statement, whatever the outcome, and quote it in the manuscript.
If the committee confirms the study was exempt, name the exemption category and the committee that granted it.
If approval was granted late, state the date and explain the circumstances in the cover letter.
If no committee exists at your institution, say so and name the national or regional body you approached instead.
Never state that approval was obtained when only a waiver or exemption was issued; the 2 are different.

Ethical Declarations Outside Biomedicine

Most published guidance is written for clinical research, which leaves authors in other fields guessing. The obligations still apply; the wording changes.

Social sciences and qualitative research

Declare approval from the relevant committee, and name the professional code you followed, such as those issued by national psychological or sociological associations.
Describe the consent process, including how you handled ongoing consent in longitudinal fieldwork.
Explain anonymization of quotations, pseudonyms, and any changed identifying details, and note that you changed them.
Address power imbalances between researcher and participant, and any incentive or payment offered.
Declare reflexivity and positionality where the field expects it, and follow COREQ or SRQR for reporting.
Explain data retention, secure storage, and destruction schedules for recordings and transcripts.

Research using social media and web-scraped data

State whether the data were public, semi-public, or private, and how you decided.
Declare compliance with platform terms of service and any application programming interface license.
Explain whether consent was sought, and justify the decision if it was not.
Describe how you prevented re-identification, including whether verbatim posts are quoted, since a quotation is searchable.
Note approval from an ethics committee even where the data are public, because many committees now require review.

Computer science, engineering, and applied AI research

Provide a broader impact or societal impact statement where the venue requires one.
Declare the provenance and license of every training dataset, and whether it contains personal data.
Declare human involvement in annotation or evaluation, including crowdworker recruitment, pay rates, and ethics review.
State model release decisions, safety evaluations, and any capability withheld.
Declare compute resources and estimated environmental cost where the venue asks for it.
Follow responsible disclosure norms for any vulnerability discovered.

Do you need ethics approval for your study?

If your work involves living people, identifiable personal data, human tissue, or animals, assume you need approval or a documented exemption. Only a committee can confirm that a study is exempt; the author cannot decide alone.

Work through the questions below in order, and record the answer to each.

Does the study involve living human participants, including through interviews, surveys, experiments, or observation? If yes, approval is required.
Does it use identifiable or re-identifiable personal data, including clinical records, images, or genomic data? If yes, approval or a documented waiver is required.
Does it use human tissue, cell lines of human origin, or biobank samples? If yes, approval and provenance documentation are required.
Does it involve live vertebrates or cephalopods? If yes, animal ethics approval is required.
Is it a secondary analysis of a fully anonymized public dataset? Approval may not be needed, but ask the committee for written confirmation.
Is it a service evaluation, audit, or quality improvement project? Local rules vary widely; obtain a written determination.
Does it involve vulnerable groups, deception, or sensitive topics? Expect full committee review rather than expedited review.
Is it a multi-country study? Obtain approval in every country where data are collected, not only at the lead institution.

Where Does Each Declaration Go in Your Submission?

The same information is often required in more than 1 place, and the copies must match. Editorial staff compare them.

Declaration	Usual location	Common mistake
Ethics approval	Methods section, plus the end matter and the submission form.	Approval number appears in the form but not in the Methods.
Consent to participate	Methods section and end matter.	Mentioned only in the cover letter.
Consent for publication	End matter; signed form retained by the author.	Sending the signed form as a public supplementary file.
Author contributions	End matter, entered as CRediT roles in the submission system.	Roles listed do not match the author order or the number of authors.
Competing interests	End matter, plus the ICMJE disclosure form for each author.	Only the corresponding author completes the form.
Funding	End matter and the funder fields in the submission system.	Grant number missing, or the funder name does not match the registry.
AI use disclosure	Cover letter and a named section in the manuscript.	Disclosed in the cover letter only, which the 2026 rules do not allow.
Data availability	End matter, before the references.	Placed in the supplementary material where indexers cannot find it.
Trial registration	End of the abstract, and repeated in the Methods.	Given only in the Methods, so it is invisible in databases.
Prior publication and preprints	Cover letter and the submission form.	Preprint not declared, then flagged by similarity screening.
Permissions and credit lines	Figure and table captions.	Credit line reworded rather than reproduced exactly as licensed.
Acknowledgments	End matter, before the references.	Naming people who did not give permission.

How do publisher requirements differ?

Core expectations are converging, but the wording, the forms, and the strictness differ. The tables below summarize general publisher positions as a starting point. Always read the instructions for the specific journal, because individual titles within a publisher can be stricter.

Requirement	Elsevier	Springer Nature	Wiley
Competing interests	Declaration of interest statement required; ICMJE form for medical titles.	Required in a Declarations section for all research articles.	Required; ICMJE form used across health science titles.
Data availability	Required or strongly encouraged, with a research data policy tier per journal.	Mandatory statement in most research journals, with 4 policy types.	Required or encouraged depending on the journal policy tier.
AI use disclosure	Declaration required for generative AI use in writing; AI cannot be an author.	Disclosure required in Methods or acknowledgments; AI cannot be an author.	Disclosure required in the manuscript; AI cannot be an author.
CRediT contributions	Supported and encouraged, required by many titles.	Encouraged, required by a growing number of titles.	Supported and encouraged across most journals.
ORCID	Required for the corresponding author at most journals.	Required for the corresponding author at most journals.	Required or strongly encouraged for the corresponding author.
Preprints	Accepted at most journals; must be declared at submission.	Preprint posting supported; must be declared.	Accepted at most journals; must be declared.

Requirement	Taylor and Francis	SAGE	PLOS
Competing interests	Disclosure statement required in every article.	Declaration of conflicting interests required in every article.	Detailed competing interests statement required at submission.
Data availability	Data availability statement required or encouraged by journal.	Statement required at most journals.	Mandatory; data must be in the paper, a repository, or covered by a stated restriction.
AI use disclosure	Required; AI cannot be an author.	Required; AI cannot be an author.	Required in the Methods or acknowledgments; AI cannot be an author.
CRediT contributions	Supported; required by many titles.	Supported across the portfolio.	Required for every submission.
ORCID	Encouraged, required by some titles.	Encouraged, required by some titles.	Required for the corresponding author.
Preprints	Accepted; declaration required.	Accepted; declaration required.	Encouraged, with direct posting options.

Requirement	BMJ	Frontiers	IEEE
Competing interests	ICMJE form required from every author.	Conflict of interest statement required in every article.	Disclosure required, with emphasis on financial and organizational interests.
Data availability	Data sharing statement mandatory for research articles.	Data availability statement mandatory.	Encouraged, with a data availability statement option and IEEE DataPort.
AI use disclosure	Required; AI cannot be an author.	Required in a dedicated section; AI cannot be an author.	Required in the acknowledgments; AI cannot be an author.
CRediT contributions	Contributorship statement required, plus a named guarantor.	Author contributions statement required.	Author contributions encouraged.
ORCID	Required for the corresponding author.	Required for all authors.	Encouraged for all authors.
Preprints	Accepted; declaration required.	Accepted; declaration required.	Accepted under the IEEE preprint policy; declaration required.

Also check the patient and public involvement statement required by several health journals, the guarantor requirement used by BMJ titles, and any transparency declaration in which the lead author confirms the report is honest, accurate, and complete.

Copy-and-Paste Declaration Templates

Replace every bracketed placeholder with your own details, and delete any template you do not need. Check the journal instructions for required section headings, because the labels vary.

Ethics approval, granted

Ethics approval: This study was approved by the [name of committee or institutional review board], [institution], on [date] under reference number [number]. All procedures were carried out in accordance with the ethical standards of the institutional research committee and with the 2024 revision of the Declaration of Helsinki.

Ethics approval, waived or exempt

Ethics approval: The [name of committee], [institution], reviewed this study on [date] and determined that formal ethical approval was not required because [reason, for example: the study used fully anonymized existing data]. The determination is recorded under reference number [number].

Informed consent to participate

Consent to participate: Written informed consent was obtained from all individual participants included in the study. For participants under 18 years of age, written informed consent was obtained from a parent or legal guardian, and assent was obtained from the participant.

Consent for publication

Consent for publication: Written informed consent for publication of the clinical details and images in this article was obtained from the participant. A copy of the consent form is held by the corresponding author and is available for review by the Editor on request.

Competing interests, present

Competing interests: [Author name] has received consultancy fees from [company] and research funding from [company] within the past 36 months. [Author name] holds patent [number] related to [subject]. The remaining authors declare that they have no competing interests.

Competing interests, none

Competing interests: All authors declare that they have no financial, personal, or professional competing interests relating to the content of this article.

Funding, with support

Funding: This work was supported by [funder name, as listed in the Crossref Funder Registry] under grant number [number] awarded to [author initials]. The funder had no role in the study design, data collection, data analysis, interpretation of the results, preparation of the manuscript, or the decision to submit it for publication.

Funding, none

Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

AI use, tools used

Use of artificial intelligence: During the preparation of this manuscript, the authors used [tool name and version] between [start date] and [end date] for [specific task, for example: language editing of the Discussion section]. The authors reviewed and edited all output, verified every factual statement and reference, and take full responsibility for the content of the published article. No artificial intelligence tool is listed as an author, and no image in this article was generated or altered using artificial intelligence.

AI use, no tools used

Use of artificial intelligence: The authors declare that no artificial intelligence or AI-assisted technologies were used in the conception, drafting, editing, analysis, or image preparation of this manuscript.

Data availability, public repository

Data availability: The datasets generated and analyzed during this study are openly available in [repository name] at [DOI or accession number], released under a [license] license. Metadata and a data dictionary are included in the deposit.

Data availability, restricted access

Data availability: The data supporting the findings of this study contain information that could compromise participant privacy and are therefore not publicly available. De-identified data may be made available to qualified researchers for [permitted purposes] following approval of a written request by [named committee or office] at [institutional contact route]. Requests are usually answered within [number] weeks.

Data availability, no new data

Data availability: No new data were created or analyzed in this study. All sources analyzed are cited in the reference list.

Code availability

Code availability: The analysis code used in this study is available at [repository URL], archived at [DOI] under a [license] license. Analyses were run using [software and version] with packages [names and versions].

Clinical trial registration

Trial registration: [Registry name], [registration number]. Registered [date], prospectively, before enrollment of the first participant. The full protocol and statistical analysis plan are available at [URL].

Prior publication and preprint

Prior publication: This manuscript is not under consideration at any other journal. An earlier version was posted as a preprint on [server] on [date] and is available at [DOI]. Preliminary findings were presented as a poster at [conference], [location], [date]. No part of this work has been published in a peer-reviewed journal.

Author contributions using CRediT

Author contributions: [Initials]: Conceptualization, Methodology, Writing (original draft). [Initials]: Investigation, Data curation, Formal analysis. [Initials]: Validation, Writing (review and editing). [Initials]: Supervision, Funding acquisition. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work.

Permissions and reuse

Permissions: Figure [number] is reproduced from [author, title, journal, year, DOI] with permission from [rights holder], license number [number]. Table [number] is adapted from [source] under a [license] license.

Inclusion and ethics in global research

Inclusion and ethics: This study was designed and conducted in partnership with researchers based in [country], who contributed to the study design, data collection, analysis, and authorship. Ethical approval was obtained from [local committee] in addition to [home institution committee]. Findings will be shared with participating communities through [route]. Local and regional literature has been cited throughout.

Animal research

Animal ethics: All procedures involving animals were approved by the [committee name] at [institution] under protocol number [number] and were carried out in accordance with [legislation]. The study is reported in line with the ARRIVE 2.0 guidelines, and the completed checklist is submitted as a supplementary file. The 3Rs were applied through [brief description].

Acknowledgments

Acknowledgments: The authors thank [name] of [institution] for [contribution]. Medical writing support was provided by [name] of [organization] and was funded by [sponsor]. All named individuals have given written permission to be acknowledged.

Pre-Submission Ethics Checklist

Work through this checklist before you upload anything. Most items cannot be fixed retrospectively.

Before you collect data

Ethics approval obtained in writing, with the number and date recorded.
Approval covers every site, every country, and every dataset in the study.
Clinical trial or systematic review registered prospectively, with the number saved.
Consent forms approved, translated where needed, and version-controlled.
Data protection assessment completed and the lawful basis for processing documented.
Authorship expectations agreed in writing across the team.
Data sharing and code sharing plans agreed with collaborators and funders.

While you write

The matching EQUATOR reporting checklist is downloaded and completed.
Every author meets all 4 ICMJE criteria; contributors who do not are moved to the acknowledgments.
CRediT roles are assigned for every author.
Permission secured and logged for every reused figure, table, or instrument.
Original unprocessed image files are archived and accessible.
Every AI-assisted task is logged with the tool, version, and date.
The Declaration of Helsinki is cited in its 2024 version where relevant.

Before you click submit

All required declarations are present: ethics, consent, publication consent, authorship, competing interests, funding, AI use, data availability, code availability, registration, prior publication, permissions, and acknowledgments.
No declaration section is left blank; each says either what applies or that nothing applies.
The submission form, the cover letter, and the manuscript say the same thing.
The trial registration number appears at the end of the abstract.
Every author has completed an individual disclosure form where the journal requires one.
ORCID identifiers are attached for the authors the journal requires.
A similarity check has been run, and any overlap with a preprint or thesis is explained.
Everyone named in the acknowledgments has agreed in writing.
The journal instructions for authors have been read in full, in their current version.

What happens if a declaration is missing or wrong after publication?

Journals respond with a correction, an expression of concern, or a retraction, depending on how serious the problem is. COPE flowcharts guide editors through each case, and the outcome stays attached to the article permanently.

Outcome	Typical trigger	Effect on the article
Corrigendum or erratum	An omitted funder, a missing competing interest, or an author list error.	The article stands; a linked notice records the change.
Expression of concern	A credible but unresolved integrity question, or an ongoing institutional inquiry.	The article stands with a public warning attached while the case is investigated.
Retraction	Fabrication, falsification, plagiarism, missing ethics approval, or paper mill involvement.	The article is withdrawn from the record, and the notice remains permanently visible.
Institutional referral	Suspected misconduct that the journal cannot resolve alone.	The employer investigates; consequences can extend beyond the single paper.

The reputational cost falls on the whole author list, not only the person responsible. Retracted work continues to be indexed and cited, and funders and promotion committees check retraction records. Disclosing a problem yourself, promptly, almost always produces a better outcome than waiting to be found.

Frequently Asked Questions

Do I need ethics approval for a retrospective chart review?

Usually yes, or at least a documented waiver. Most committees treat access to identifiable medical records as human subjects research even when no patient is contacted, and journals ask for the reference number.

Apply for review before you extract any data. If your committee grants a consent waiver rather than full approval, name the waiver and the committee in the manuscript, and do not describe it as approval. If the committee rules the study exempt, ask for that ruling in writing so you can quote it.

How do I write a conflict of interest statement when I have nothing to declare?

Write it explicitly rather than leaving the section blank. A sentence stating that all authors declare no financial, personal, or professional competing interests relating to the article is what editors expect to see.

Confirm with every co-author first, and check the full 36-month look-back period. Speaking fees, unpaid advisory roles, editorial positions at the journal, and patent applications are all disclosable, and authors frequently forget them.

Can I submit my paper to 2 journals at the same time?

No. Simultaneous submission is prohibited by essentially every peer-reviewed journal, and being caught usually means rejection at both titles and a note on your record.

Wait for a decision before you submit elsewhere. If a journal is unacceptably slow, withdraw formally in writing and confirm the withdrawal has been processed before submitting to the next journal. Posting a preprint is a legitimate way to establish priority while you wait.

Do I need to declare using ChatGPT for grammar checking?

Many journals exempt AI for basic grammar checks from disclosure. However, you need to be very sure that you used the tool only for grammar, punctuation, or spelling errors and didn’t use any of its generative AI features. If you’re not sure about the difference between “editing suggestions” and “new wording” from the tool, disclose its use.

Name the tool and version, state which sections you used it on, and confirm that you reviewed and verified the output. Over-disclosure carries no penalty; non-disclosure can be treated as misconduct.

Is posting a preprint considered prior publication?

At most journals, no. A preprint is generally not treated as prior publication, but you must declare it at submission with the server, DOI, and posting date.

Check the specific journal before posting, because a minority of titles, particularly in some clinical specialties, still decline preprinted submissions. Also check the license you applied, since some licenses can complicate the transfer of rights later. If you update the preprint after acceptance, follow the journal policy on which version may be posted.

What is the difference between a funding statement and a conflict of interest statement?

A funding statement records who paid for the research and what role they played. A conflict of interest statement records relationships that could bias the work, including but not limited to funding.

The same grant can appear in both. Funding from a company whose product you evaluated belongs in the funding statement as support, and in the competing interests statement as a relationship that a reader might consider relevant. Listing it in only 1 place is a common reason for editorial queries.

Do case reports need ethics approval and patient consent?

Consent for publication is always required. Formal ethics approval is often not, because a single case report is usually not classed as research, but many institutions still want a written determination.

Obtain written consent from the patient on the journal form, show them exactly what will be published, and retain the signed form rather than uploading it. Anonymize dates, locations, occupations, and rare details that could identify the person. Follow the CARE reporting guidelines, and check the journal policy for cases where the patient has died.

What should I write in a data availability statement if my data cannot be shared?

State the restriction, name the authority that imposed it, and give a real route to controlled access. A statement that simply says data are available on request is increasingly rejected by editors.

Identify the ethics committee, law, or agreement preventing open sharing. Name an institutional office or data access committee rather than a personal email address, describe the conditions applicants must meet, and give an expected response time. Where possible, share a de-identified subset, aggregated results, or synthetic data so that some verification is possible.

This article was originally published on September 19, 2022, and updated on August 4, 2026.

Ethical Declarations for Journal Submission: Sample Statements and Publisher Requirements