Living Systematic Review Authorship

A living systematic review (LSR) is not a paper that gets published once. It is a review that is searched, screened, and reanalyzed on a recurring schedule — continuously, monthly, or on a defined trigger — for as long as new evidence keeps changing the answer to a clinically or policy-relevant question. That single design choice breaks the assumption every other authorship model relies on: that a byline is decided once, at submission, for a fixed piece of work. An LSR has no single submission moment. It has a sequence of them, and the team doing the work at update 6 may look nothing like the team that wrote the original protocol. This guide covers how authorship and contributorship are actually handled when the underlying publication itself is designed to keep changing.

Why living systematic review authorship is a different problem

A conventional systematic review is credited the way any multi-author manuscript is: contributors are assessed against authorship criteria once, a byline is fixed, and that byline attaches permanently to one version of record. An living systematic review (see also the related concept of a living review) instead produces a sequence of version-of-record updates over months or years, each one a distinct publication event with its own search, screening, and synthesis work. Cochrane, which has led formal adoption of the LSR model, and the Journal of Clinical Epidemiology’s founding LSR methods series (Elliott et al., ‘Living systematic review: 1. Introduction — the why, what, when, and how,’ 2017) both treat authorship, workload, and team sustainability as first-order design problems for an LSR, not an afterthought — because a review that never stops updating needs a team structure that can survive turnover the original one-off review never had to plan for.

Two structural facts drive everything else in this guide:

Contribution is time-bound. Someone who ran the search and did dual screening for the 2023 update may have no involvement at all in the 2025 update, and vice versa.
The ‘paper’ is a moving target. Citation, credit, and version history all have to track which specific update a person actually worked on, not just ‘the review’ as an undifferentiated whole.

The baseline doesn’t change: ICMJE criteria still apply, reassessed at every update

Nothing about the living format loosens who counts as an author. Cochrane’s own LSR guidance requires author teams to follow standard ICMJE authorship and conflict-of-interest policy at every stage of an LSR’s life — protocol, new review, each updated review, and any accompanying editorial — and to revisit team composition as the work progresses rather than treating the original byline as fixed for the life of the review. The mechanism practitioners actually use, per a 2023 scoping review of LSR conduct and reporting guidance, is straightforward in principle even though detailed frameworks for it are still sparse in the published literature: authorship is reassessed for each new version, and each team member’s contribution to that specific version is checked against the standard ICMJE four-part authorship criteria — substantial contribution to conception/design or data acquisition/analysis/interpretation, drafting or critically revising the work, final approval of the version being published, and agreement to be accountable for it.

In practice this means the question an LSR team has to answer isn’t ‘was this person an author of the review’ — it’s ‘did this person meet authorship criteria for THIS update.’ Someone can legitimately be an author on versions 2 and 4 and not on version 3, if that’s genuinely where their contribution fell.

Core team, rotating contributors, and the guarantor role

Most functioning LSR teams split into two tiers, even where no formal policy names them that way:

A core or maintaining team — typically including a guarantor or corresponding-author role — that persists across cycles, holds institutional memory of prior search strategies and inclusion decisions, and is responsible for deciding when a trigger for updating has been met.
Rotating or occasional contributors — a second screener brought in for one update’s volume, a statistician consulted when a new comparator needs a fresh meta-analysis, a topic expert added for a single evidence surge — whose involvement is real, citable, and update-specific rather than continuous.

Cochrane’s guidance and the practical literature on sustaining an LSR (including a 2023 practical guide on updating a living review published in the journal Systematic Reviews, DOI 10.1186/s13643-023-02325-y, written from experience keeping COVID-19-era LSRs current) both flag workload and team sustainability as the operational risk that most often kills a living review outright, separate from the authorship-accounting question. A team structure that assumes every past contributor is available and willing to keep re-approving future versions doesn’t survive contact with a multi-year update schedule.

Where the field’s guidance actually is — and where it genuinely isn’t

It’s worth being direct about this rather than implying a fully solved practice exists: a 2023 scoping review of methods and guidance for conducting, reporting, publishing, and appraising living systematic reviews found that only a minority of the guidance documents it examined addressed authorship-turnover across updates at all, and that the papers which did largely stopped at ‘reassess against ICMJE criteria for each version’ without offering a detailed operational framework for core-versus-rotating team structures, contributor logging, or transparent disclosure of who did what on which update. That gap is exactly the space this guide’s practical framework below is meant to fill with a synthesis of what the standards that DO exist — ICMJE criteria and the CRediT taxonomy — actually imply when applied to a multi-version publication, not a claim that a single authoritative LSR-specific standard already covers it.

CRediT: the mechanism that actually fits a versioned publication

The CRediT taxonomy (formalized as ANSI/NISO Z39.104-2022) was built to answer ‘who did what,’ role by role, rather than ‘who is the author, ranked first to last.’ That role-based structure fits an LSR’s version-by-version reality far better than a single fixed byline does, because it’s already designed to record contribution at a granularity finer than a whole paper: a contributor’s Investigation, Formal Analysis, or Writing – Review & Editing role can be logged separately for each update they actually touched. A practical version of this for an LSR team is a per-update contribution log: for every update cycle, record which CRediT roles each named contributor performed for THAT update specifically, alongside the ICMJE authorship determination for that version. Over several update cycles this produces an auditable history of who is creditable for what, rather than a single static byline that either overclaims (crediting someone for work they didn’t do on a later version) or underclaims (silently dropping someone whose earlier contribution is real and citable).

What happens to a contributor who no longer qualifies

Turnover is normal in a multi-year living review, and the standard authorship mechanism for handling it is not unique to LSRs — it’s the same rule ICMJE applies to any paper: once a person’s contribution to the current version genuinely doesn’t meet authorship criteria, they’re not an author of that version. That is not a demotion or a judgment on their earlier work; it’s a description of that specific update. The corresponding practice, consistent with standard authorship guidance, is to move them into an acknowledgments section for that version (crediting their prior role explicitly, e.g. ‘X contributed to the search strategy for the 2022 and 2023 updates’) rather than silently dropping their name with no record at all. Two things materially reduce the risk of this becoming an authorship dispute later:

Setting authorship eligibility criteria for the review — including how eligibility will be reassessed at each update — at team formation, before anyone’s inclusion is in question, per Cochrane’s guidance to define and revisit authorship policy from the review’s inception.
Communicating a reassessment decision to the affected contributor directly and with the specific reasoning, rather than letting them discover the change by reading the published update. Guidance on mediating authorship disputes applies here as much as it does to any other authorship disagreement — most disputes trace back to a decision made without transparent criteria or timely communication, not to the decision itself being wrong.

Byline structure and citation practice across versions

Because an LSR’s contributor list can differ meaningfully version to version, teams generally handle the visible byline one of two ways: (1) a versioned individual byline, where each published update carries its own author list reflecting that update’s contributors, cross-referenced to a stable review identifier and a change log describing what’s new since the prior version; or (2) a standing group or consortium byline for the review as an ongoing entity (analogous to the collective-authorship conventions used by very large collaborations — see how megacollaboration authorship works in fields like high-energy physics for a related but distinct large-team model), with individual contribution detail carried in a per-update CRediT statement rather than in the citable byline itself. Either approach depends on the same underlying discipline: a clear, dated ‘what’s new in this version’ note and a contribution record that ties specific people to the specific update they worked on. Readers and citing authors need to be able to tell which version they’re citing and, if they need to, which team produced it — the same version-tracking discipline covered in general terms in this site’s guide to managing manuscript versions across co-authors, applied here to a publication that keeps versioning indefinitely rather than converging on one final manuscript.

A practical framework for LSR teams

Synthesizing Cochrane’s LSR guidance, standard ICMJE criteria, and the CRediT taxonomy into an operational checklist for a team running (or sponsoring) a living systematic review:

Set authorship policy at inception, not at the first dispute. Define eligibility criteria and the reassessment process for every future update before the first version is even published.
Reassess authorship at every update trigger, applying the standard ICMJE criteria to that version’s actual work, not to historical involvement in the review generally.
Log CRediT roles per update, not just once for the review as a whole — this is what makes a fair authorship reassessment auditable later.
Designate a persistent guarantor or corresponding-author role that survives contributor turnover and owns update-trigger decisions, search-strategy continuity, and version-history integrity.
Move non-qualifying past contributors to acknowledgments with an explicit, specific description of what they contributed and when — never a silent removal.
Publish a dated change log with every update stating what changed in the evidence, the methods, and the author team, so citing readers can see exactly what they’re citing.
Keep a contribution record contributors can cite on a CV or in a promotion/RCR file that specifies which update(s) they actually worked on and in what CRediT role — this matters for researcher career documentation in a way a single undifferentiated ‘author of the living review’ credit does not.

Frequently asked questions

Does everyone stay listed as an author every time a living systematic review is updated?

No. Authorship is meant to be reassessed at each update against standard ICMJE criteria applied to that specific version’s work. A contributor who doesn’t meet the criteria for a given update is not an author of that update, though their earlier contribution should still be documented, typically in acknowledgments.

How is this different from authorship on an ordinary updated systematic review?

A conventionally updated systematic review (a new edition published occasionally, with its own submission and byline decided at that point) still ends up with one fixed byline per version, decided once. A living systematic review is designed around a recurring, often frequent, update schedule from the outset, which makes authorship turnover a routine operational question rather than an occasional one — and makes a structured, per-version reassessment process (versus a one-off byline decision) the practical necessity rather than the exception.

Can someone who only worked on one update still claim authorship credit for the review?

Yes, for the specific version they contributed to. Their contribution record should specify which update(s) they worked on and what CRediT role(s) they performed, rather than an undifferentiated claim to authorship of ‘the living review’ as a whole spanning versions they had no part in.

Should a living review team use CRediT roles, ICMJE criteria, or both?

Both, and they serve different jobs. ICMJE’s four-part criteria answer the binary question of who qualifies as an author of a given version. CRediT answers the finer-grained question of what each contributor actually did, which is what makes a fair, auditable authorship reassessment possible in the first place and what supports acknowledgment of non-author contributors.

How should a living systematic review with multiple past author teams be cited?

Cite the specific version consulted, using that version’s own byline and publication date, along with the review’s stable identifier if one exists. This is the same discipline as citing any versioned publication — the change log accompanying each update should make clear which version corresponds to which author team.

Living Systematic Review Authorship: How Rotating Contributor Teams Get Credited