Large multicenter clinical trials routinely involve dozens or hundreds of participating sites, each contributing investigators, coordinators, and staff. When the trial’s primary results are published, the byline cannot practically name every one of those people as an individual author — but not everyone who touches a trial has done work that meets authorship criteria in the first place. Trials solve this with a two-tier structure: a small, named writing committee credited as the visible authors (often under a collaborative group byline, such as “on behalf of the [Trial Name] Investigators”), and a much larger full investigator/collaborator list published separately, typically in a supplementary appendix. This guide explains how that split works, who belongs in each tier, and how it interacts with ICMJE’s authorship criteria, PubMed/MEDLINE indexing, and CRediT contribution statements. For the general mechanics of naming a collaboration as an author entity, see CASRAI’s Group Authorship entry; this guide focuses specifically on how that pattern plays out in clinical trial publications.
Two tiers, one trial: writing committee and full list
ICMJE’s guidance on authors and contributors describes group authorship as occurring in two configurations: a group name published alongside every individual author’s name, or a group name as the primary visible byline with the full roster of individuals listed separately. Large clinical trials almost always use the second pattern. The published article’s byline names the trial’s collaborative group (for example, “The RECOVERY Collaborative Group”) together with a small, individually credited writing committee — typically somewhere between half a dozen and twenty named people — who are listed as the visible authors of record. The much longer roster of participating site investigators, sub-investigators, coordinators, and other collaborators who do not meet full authorship criteria is published separately, usually as a supplementary appendix to the article rather than in the printed byline itself.
The UK’s RECOVERY platform trial is a real, widely cited example of this configuration: its primary COVID-19 treatment results, including the dexamethasone findings published in the New England Journal of Medicine, appeared under the byline “The RECOVERY Collaborative Group,” with a small named writing committee credited individually as authors and the full list of site investigators and collaborators published separately as supplementary material.
Writing committee membership still has to clear the ICMJE bar
The critical point — and the one most often misunderstood on large trials — is that being named to the writing committee is not a courtesy or a seniority marker. ICMJE is explicit that a group byline does not relax the underlying authorship test: every individual credited as an author, whether visible on the byline or listed in a group’s separate roster, must independently satisfy all four of ICMJE’s authorship criteria (substantial contribution to the work; drafting or critically revising it for important intellectual content; final approval of the version published; and agreement to be accountable for the accuracy and integrity of the whole work). See CASRAI’s ICMJE Authorship Criteria guide for the full four-part test. On a trial with hundreds of contributing sites, most site investigators contribute substantially to data acquisition — one of the four criteria — but relatively few also draft or critically revise the manuscript and take accountability for the whole work. That narrower group is who ends up on the writing committee; the rest are appropriately credited as collaborators rather than authors, which is a description of their actual role, not a demotion.
How trials decide who is on the writing committee
Because a large trial’s authorship stakes are high and the pool of eligible contributors is enormous, well-run trials do not leave writing-committee selection to be worked out informally after data lock. The trial’s own publication policy — ideally documented in the protocol or a separate authorship charter agreed before enrollment begins — typically specifies how writing committee members are chosen. Common approaches include naming the trial’s steering committee, lead statistician, and principal investigators from the highest-enrolling or methodologically central sites as the default writing group, sometimes supplemented by rotating individual site investigators into later or secondary publications from the same trial. Some large trials formalize this further with an explicit, criteria-based nomination process: the ISCHEMIA trial, for example, used a published point-based authorship system in which participating sites accrued points from objective performance measures — enrollment volume, randomization rate, data completeness and timeliness, and protocol adherence — which were then used to allocate a defined number of named authorship positions, while the remaining participating investigators were recognized as collaborators through a comprehensive list submitted for PubMed indexing. Deciding this mechanism up front, rather than after the primary results are drafted, is the single most effective way to avoid the authorship disputes that under-documented large-group projects are prone to; see CASRAI’s guide to authorship agreements for the general principle and how authorship disputes get resolved if one arises anyway.
What happens to everyone who isn’t on the writing committee
Being left off the writing committee does not mean a contributor goes uncredited. Per ICMJE’s guidance on how MEDLINE handles group-authored papers, when a byline includes a group name, MEDLINE lists the individual group members separately, tagging each as either an author or a collaborator based on what the journal supplies. In practice this means the full investigator list for a trial — often numbering in the hundreds — is submitted alongside the manuscript and indexed in the PubMed record even though those names do not appear in the article’s short citation form. This is why searching PubMed for a specific site investigator’s name can surface a trial publication that doesn’t name them anywhere in its printed citation: they are indexed as a collaborator on the record itself. Journals typically also require the full list to appear somewhere in the published article — most commonly a numbered supplementary appendix organized by site — so the contribution remains a matter of public, citable record even though it is not part of the author byline.
Writing committee vs. steering committee vs. full author list
These three groups are often confused because their membership overlaps heavily, but they answer different questions. A trial’s steering committee is a governance body: it oversees trial conduct, protocol amendments, and major operational decisions, and its remit exists independently of who ends up writing any given manuscript. The writing committee is specific to a publication: it is the group of people who actually draft and take accountability for a particular paper, and a trial with several major publications (primary outcome, subgroup analyses, secondary endpoints) may assemble a different writing committee, drawn from a different subset of eligible contributors, for each one. The full investigator/collaborator list is broader still — it includes everyone who contributed to running the trial at any site, most of whom meet the data-acquisition criterion but not the drafting/revision and accountability criteria that would make them individual authors. A steering committee member is very often also on the writing committee, but the two roles are not the same thing, and neither guarantees authorship on any specific paper by itself.
CRediT contribution statements alongside a group byline
Trials that also publish a CRediT contribution statement face an additional wrinkle: CRediT was designed around individually named authors, so a group byline needs the statement to map roles to the writing committee members by name (or, less commonly, to defined sub-groups such as “statistical analysis team”) rather than to the collaborative group as a whole. Journals that require both a group byline and a CRediT statement generally expect the statement to cover the named writing committee only, with the trial’s protocol-stage authorship policy serving as the record of how the broader contributor pool’s work was recognized instead.
Practical guidance for trial coordinating centers and research administrators
- Write the authorship policy into the protocol, not after data lock. Specify how the writing committee will be selected, what role (steering committee, site performance metrics, statistical leadership) determines eligibility, and how later secondary publications from the same trial will rotate credit.
- Apply the ICMJE four-criteria test to every proposed writing committee member individually — group or committee membership is not itself a substitute for meeting the criteria.
- Collect the full collaborator list early and keep it current. Site investigator rosters change over a multi-year trial; reconciling names, affiliations, and consent to be listed at the point of submission is a real administrative burden that is much easier to manage as a running list than as a last-minute compilation.
- Confirm the target journal’s specific submission mechanics for group authorship (how the collaborator appendix should be formatted, whether a CRediT statement is required, how the group name itself should be worded) before drafting — these details vary by journal even though the underlying ICMJE principles are shared.
- Keep documentation of the selection process. If a writing-committee decision is ever challenged, a documented, criteria-based selection process is the difference between a quick resolution and a protracted dispute.
Frequently asked questions
Who should be on a clinical trial’s writing committee?
Whoever can independently satisfy all four ICMJE authorship criteria for that specific manuscript — substantial contribution, drafting or critical revision, final approval, and accountability for the whole work. In practice this is usually drawn from the trial’s steering committee, lead statistician, and a limited number of principal investigators, selected according to a policy the trial ideally documents before enrollment begins rather than worked out informally once results are in.
Does every site investigator on a trial count as an author?
No. Most site investigators contribute substantially to data acquisition, which is one of ICMJE’s four criteria, but that alone does not meet the full bar. Investigators who don’t also draft or critically revise the manuscript and accept accountability for it are appropriately credited as collaborators, not authors — typically via a supplementary investigator list that is still indexed in PubMed.
Is a writing committee the same as a steering committee?
No, though membership often overlaps. A steering committee is a governance body overseeing the trial’s conduct; a writing committee is specific to a given manuscript and can differ across the trial’s various publications.
How does the full investigator list get credited if it isn’t in the byline?
Per ICMJE’s guidance on MEDLINE indexing, journals submit the full roster alongside the manuscript, and MEDLINE lists each individual separately, tagged as either author or collaborator. The list is also typically published as a numbered supplementary appendix to the article itself.
Does a group byline lower the bar for authorship?
No. ICMJE is explicit that using a group or collaboration name on the byline does not relax individual accountability — every person credited as an author within the group, whether named on the byline or in a separate list, must independently meet all four authorship criteria.
For the broader mechanics of naming conventions and how large collaborations (including non-clinical examples like physics megacollaborations) structure author bylines, see CASRAI’s Group Authorship entry and the Clinical Research Administration pillar for the wider set of clinical trial operations topics. For how authorship order is conventionally decided once the writing committee is set, see How Authorship Order Is Decided, and for the protocol-stage planning that should include an authorship policy, see Designing a Clinical Trial Protocol.
