23. Authorship, Publication & Knowledge Sharing

ICMJE authorship: four prongs

The International Committee of Medical Journal Editors (ICMJE) authorship criteria are the most widely adopted standard:

1. Substantial contribution to conception, design, acquisition, analysis, or interpretation of data
2. Drafting or revising the work critically for important intellectual content
3. Final approval of the version to be published
4. Accountability for all aspects of the work, with the ability to ensure questions about accuracy or integrity are appropriately addressed

All four. Not three. Not "substantial contribution alone." A person who provided lab space, helped recruit, or read drafts without proposing changes may merit acknowledgment, not authorship.

Why this matters: authorship signals accountability. A study with seven authors is making a claim about who can be held responsible for its content. Inflating the author list erodes the meaning of authorship across the field.

Honorary authorship

Honorary authorship is including someone as an author who didn't meet the criteria — most often a senior figure whose name is added for prestige, funding, or political support. It violates ICMJE criteria, distorts credit, and is structurally common.

Practical defense:

• An authorship discussion before drafting starts
• Documented contributions for each author, mapped against ICMJE criteria
• A norm that "I helped recruit" or "I gave you bench space" warrants acknowledgment, not authorship
• Senior researchers actively declining honorary positions

The norm shifts when senior researchers refuse honorary credit, not when junior researchers refuse to grant it.

Ghost authorship

Ghost authorship is the omission of someone who did meet criteria — most often industry-funded ghost-writers, but also sometimes community partners or junior collaborators whose contributions are absorbed into senior authors' credit.

Equally distorting; equally important to avoid. The fix is the same: explicit pre-drafting authorship discussions with documented contributions.

Authorship in community-engaged research

Community partners who contributed to design, data collection, or interpretation likely meet ICMJE criteria. Standard practice has often relegated them to acknowledgments. Better practice:

• Authorship offered to community partners who meet criteria
• Some community partners may decline authorship — for confidentiality, capacity, or organizational reasons; respect the decline
• For those who accept, support the practical work of being an author (review drafts, sign off on submissions)
• Where authorship is declined, formal acknowledgment with explicit description of contribution

The work is to make authorship offers real, not perfunctory. A community partner who's offered authorship in an unintelligible draft three days before submission isn't being offered meaningful co-authorship.

Authorship order

Authorship order conventions vary by field. Common patterns:

• First author — primary intellectual contributor; usually the person who did the work
• Last author / senior author — group leader or PI; intellectual oversight responsibility
• Middle authors — significant contributors ordered variously (by contribution, alphabetical, or in conventions specific to the field)
• Corresponding author — responsible for communication with the journal; usually first or last

When community partners are authors, the order should reflect contribution honestly. A community organization PI who held substantive intellectual responsibility could appropriately be a senior or corresponding author.

Preregistration

Preregistration is the practice of filing a study plan publicly before data collection (or before analysis, for secondary-data studies). It includes hypotheses, methods, sample, and analytic plan.

Benefits:

• Distinguishes confirmatory from exploratory analyses
• Limits researcher degrees of freedom that lead to spurious findings
• Provides a public record of pre-data-collection intentions
• Strengthens reviewers' ability to assess the study

For quantitative studies with clear hypotheses, preregistration is increasingly an expectation. Tools: OSF, AsPredicted, Clinicaltrials.gov, registered reports at participating journals.

For qualitative work, the analog is protocol publication — making the design protocol public even when specific hypotheses aren't appropriate. The level of pre-specification differs; the commitment to transparency doesn't.

Open data and code

Open data: depositing anonymized data and a codebook in a public repository. Open code: posting analysis scripts in a public repository (GitHub, OSF, Zenodo).

Both are increasingly expected in many fields and many funders' policies. They support reproducibility and enable other researchers to extend or critique findings.

Constraints to acknowledge:

• Some data cannot be openly shared (sensitive populations, sovereignty commitments)
• Code may be entangled with proprietary tools
• Documentation effort is real; budget for it

When full openness isn't possible, partial openness — code without data, summary data without raw records — is usually better than nothing.

Preprints

Preprints are early versions of papers posted to public servers (medRxiv, bioRxiv, SSRN, etc.) before or alongside journal submission. Benefits:

• Faster dissemination of findings
• Public record that doesn't depend on journal acceptance
• Opportunity for community feedback before publication

Most major journals now allow preprints. Check journal policies before posting; some still don't.

Registered reports

Registered reports are a publication format where the study is reviewed and accepted in principle before data collection. The journal commits to publishing the results regardless of whether they support the hypothesis, provided the protocol was followed.

This format directly addresses publication bias and the file-drawer problem. It's a particularly powerful format for confirmatory studies. Adoption is growing across many fields.

Choosing where to publish

The journal matters. Considerations:

• Audience — who reads this journal? Are they the people who can use the findings?
• Open access — is the paper accessible to readers without institutional subscriptions?
• Editorial values — does the journal value methodological diversity, community engagement, mixed methods?
• Speed — how long from submission to publication?
• Article-processing charges — what's the cost and who pays?

The "highest impact factor" answer is rarely the right answer for transdisciplinary work whose audiences include practitioners, policymakers, and community members.

Beyond papers: knowledge mobilization outputs

Academic papers are one output type. Transdisciplinary projects often need:

• Policy briefs — short, decision-oriented documents for policymakers
• Practice guidelines — actionable summaries for practitioners
• Community reports — accessible, often visual, summaries for the studied community
• Training materials — for capacity-building in partner organizations
• Datasets and tools — for other researchers and analysts
• Plain-language summaries — for general audiences

The mix of outputs is part of dissemination planning, ideally specified at proposal time with budget and authorship implications.

A vignette

A community-academic team finishes a 3-year study of housing instability and mental health. The output plan:

• One peer-reviewed paper in a methods-friendly journal (community-engaged research)
• One policy brief delivered to the city housing authority
• One community report co-authored with the resident advisory council, distributed at three community meetings
• A dataset deposited with permissions enabling community-controlled secondary use
• Two staff trainings for partner clinics
• A preprint at submission to enable community review before publication

Authorship discussions happened at proposal time. The community-research-coordinator at the partner organization is second author on the journal paper and first author on the community report. The PI is corresponding author and last on the journal paper. Two community advisory council members declined journal authorship for personal reasons and are credited fully in acknowledgments with their explicit consent to be named.

Closing

ICMJE authorship has four prongs and all are required. Authorship discussions belong at the start, not the end. Community contributors deserve authorship consideration. Open-science practices — preregistration, open data and code, preprints, registered reports — strengthen transparency and reproducibility. Where to publish is a strategic choice that includes audience, access, and editorial values.

Next: communication strategies for diverse audiences — audience adaptation, multiliteracies, and accessibility.