Coercive Citation in Medical Peer Review: What to Do When a Reviewer Asks You to Add Citations

The scenario is familiar to anyone who has submitted papers for a while. The peer review comes back, and buried in the middle of otherwise reasonable feedback is a request to cite several additional papers. The reviewer does not say they wrote those papers. They do not explain how the references would improve the argument. The phrasing is either vague ("a more thorough literature review should include these works") or phrased as a condition ("revisions should incorporate the important contributions of [list of citations]"). You check the references. Several are from the same research group, or from the same journal, or both.

That is coercive citation. It is a recognized form of publication misconduct, addressed by the Committee on Publication Ethics (COPE), flagged by Clarivate in its Journal Citation Reports (JCR) enforcement process, and studied in multiple surveys of researchers across disciplines including medicine, life sciences, and clinical research. Understanding what it is, what it is not, and what you can do about it matters more now than it did five years ago, because the practice has become more visible as journal integrity monitoring has expanded.

How Common It Actually Is

Exact figures are hard to come by because peer review is confidential, complaints are usually private, and researchers often absorb the cost quietly rather than escalate. But the surveys that exist are consistent enough to suggest the problem is not marginal.

A 2017 study by Fong and Wilhite, published in Science, surveyed approximately 12,000 academics across economics, sociology, psychology, and business disciplines and found that 14.1 percent reported receiving coercive citation requests from editors. A separate survey at the National Institute of Environmental Health Sciences, conducted with 220 researchers, found that 22.7 percent had received peer review comments requesting them to add citations to work authored by the reviewer. Both figures likely undercount the true rate, because some recipients do not recognize the practice as misconduct, some add the citations and move on without reflecting on it, and some are reluctant to report it even anonymously.

Documented extreme cases put the more diffuse statistics in context. One editor at the Journal of Theoretical Biology was found to have asked authors, on average, to add 35 citations per manuscript, with 90 percent of those citations pointing to papers the editor had co-authored. In another case studied in the bioinformatics literature, a handling editor was shown to have requested self-citations more than 50 times more often than they requested citations to work they were not involved in, before eventually resigning from their role. These are not representative of all editors, but they illustrate how structural the problem can become when no monitoring is in place.

In medical and clinical research, there is an additional reason the practice persists: impact factors are used in promotion, tenure, grant applications, and department evaluations in ways that create real incentives to game them. A journal editor who also runs an active research program has something concrete to gain from each citation request that succeeds. That does not make the behavior acceptable, but it does explain why it is not rare.

The Four Forms the Practice Takes

Coercive citation is not one thing. It takes at least four distinct forms that authors encounter, and each has a slightly different ethical weight and a slightly different response path.

The most common form is reviewer self-citation coercion, where a peer reviewer uses their review to request citations to their own work without disclosing that connection. Because reviewer identities are typically confidential, authors often cannot verify whether the cited papers belong to the reviewer. This is the form most frequently captured in surveys, and it is what most people mean when they say coercive citation.

The second form is editorial self-citation pressure, where a handling editor, associate editor, or editor-in-chief asks authors to cite papers they authored or co-authored. This sometimes happens via the formal review system, but it also happens informally through editorial correspondence. Because editors have more formal power over acceptance decisions than reviewers do, this version carries a higher coercive load.

The third form is journal self-citation pressure, where the journal as an institution, rather than an individual, pressures authors to cite papers published in that same journal regardless of their relevance. This is sometimes explicit (instructions that require authors to demonstrate familiarity with the journal's output) and sometimes implicit (repeated revision requests asking for "more engagement with the recent literature" when the cited literature is exclusively from that journal). Clarivate monitors for this pattern in JCR-listed journals and can suppress a journal's impact factor or remove it from the JCR entirely if journal self-citation rates are anomalously high.

The fourth form is citation cartel behavior, where a network of journals or research groups trade citations with each other to boost their collective metrics. This is harder for individual authors to detect because the requested citations appear to come from a real and diverse body of work rather than from one person or one title. It tends to be identified through bibliometric analysis rather than individual reports.

Legitimate Citation Requests Versus Coercive Ones

Not every reviewer request to add citations is misconduct. Reviewers are supposed to catch gaps in the literature review. They are supposed to flag related work the authors missed. They are sometimes experts in exactly the sub-area the manuscript addresses, and a request to cite a paper they published is occasionally both relevant and appropriate. The difference between a legitimate citation request and a coercive one is not whether the reviewer wrote the suggested paper. It is whether the suggested paper is scientifically necessary to the manuscript.

A legitimate citation request will explain the relationship between the suggested paper and the manuscript. It might say that the study design directly addresses a methodological concern raised in the review, that a key claim the authors make was established by the suggested work and the omission is a factual gap, or that the cited study produced contradictory findings the authors should address. These requests can be evaluated on their merits. The suggested paper either fits the argument or it does not.

A coercive request typically does not explain the connection. It may list several papers in a single sentence without analysis. It may use vague language about comprehensiveness or scope. It may cluster in a single author or group without obvious topical justification. It may condition a positive recommendation on inclusion of the citations. And it may ask for more citations than the subject genuinely requires, often in a way that is disproportionate to the rest of the review.

What COPE Says Authors Should Do

COPE is the primary ethics body for academic publishing in most English-language medical journals, and its guidance on citation manipulation is unambiguous. COPE's position is that citation requests must always have a legitimate scientific rationale. A reviewer or editor who makes citation requests conditional on acceptance, or who suggests citations without scientific justification, is violating publication ethics. Authors are not obligated to comply with unjustified citation requests, and journals should not treat compliance with them as a condition of editorial consideration.

COPE's practical guidance for authors in this situation has three components. First, authors can decline to add citations they do not consider scientifically justified and should explain that decision clearly in their revision response letter. Second, authors can raise the issue with the editor-in-chief if they believe a reviewer or handling editor has made coercive requests. Third, authors can report suspected coercive citation to COPE itself if the journal fails to respond appropriately, though COPE is an advisory body and cannot compel journals to act.

The practical difficulty is that authors face an asymmetry of power. Declining a reviewer's citation request politely but on principled grounds can work. It can also result in a negative recommendation. Most journals do not reveal reviewer identities, which means authors cannot always know whether the request comes from someone with a direct financial stake. And raising the issue with an editor-in-chief only works when the editor-in-chief is not also the person making the coercive request, which is sometimes the situation authors face.

How to Respond to a Suspicious Citation Request

The most defensible response is also the most straightforward one. In your revision letter, address each suggested citation individually. For each paper you are adding, write a sentence in the response explaining why the citation improves the manuscript. For each paper you are declining to add, write a sentence explaining why it does not fit. This documentation strategy serves two purposes. It makes your scientific reasoning visible to the editor, and it creates a record if the situation escalates.

Avoid aggressive or accusatory language in the response letter. You are not sure who the reviewer is, and a defensive or hostile tone in a revision letter often works against authors regardless of who is right. The cleaner approach is to be matter-of-fact and specific: "We evaluated the suggested citations and were unable to identify a connection to our specific analysis of [topic]. We have therefore not included them." That sentence is professionally irreproachable. A sentence accusing the reviewer of self-promotion is not, even when the suspicion is correct.

If the reviewer's identity is clear from context, or if the request is unusually severe, consider a private note to the handling editor alongside the revision. Keep it factual. Name the specific requests that appeared unjustified and ask whether the journal has a policy on citation coercion. Most journals subscribe to COPE guidelines, and framing your concern as a policy question rather than a personal accusation is more likely to get a substantive response.

In the most serious cases, where you have documentation of an editor demanding citations to their own work as a condition of acceptance, COPE accepts formal disclosures from authors. Preparing such a disclosure requires retaining copies of the editorial correspondence, which most submission systems allow you to download. Before submitting a formal report, be certain about the facts. Misidentified coercion can damage a legitimate editorial relationship, and erroneous complaints to COPE are handled seriously.

What Publishers and Clarivate Are Doing About It

Publisher-level responses to coercive citation have become more systematic over the past few years. Elsevier, Springer Nature, and Wiley have all published explicit guidance for handling editors stating that citation requests from reviewers must have a stated scientific rationale. Several publishers have added monitoring flags to their editorial management systems that trigger a review when a reviewer or editor repeatedly requests citations to specific authors or journals. These flags do not catch everything, but they have led to documented cases of editorial board members losing their roles after patterns were identified.

Clarivate's enforcement mechanism operates at the journal level rather than the individual reviewer level. The JCR process includes monitoring for anomalous self-citation rates, and journals found to be inflating their citation scores through coercive practices can be suppressed from the JCR, which means their impact factor is withheld for that year. In more serious cases, journals can be delisted entirely. Between 2017 and 2024, Clarivate suppressed or removed dozens of journals for citation manipulation, with a meaningful share of the affected titles being in biomedical fields. Authors who have invested months in a submission to a journal that subsequently loses its JCR listing are entitled to feel aggrieved, and this is one of the reasons checking a journal's citation behavior history before submission is worthwhile.

In 2026, the bibliometric analysis of citation patterns has become more accessible to researchers. Tools that visualize citation networks can now show authors whether a journal's citation profile has a suspicious cluster, whether papers in the journal cite each other at anomalously high rates, and whether editorial board members are disproportionately represented in the journal's own reference lists. These are not definitive tests, but they are useful signals when evaluating a journal before submission.

The Impact on Research Literature

Citation coercion does damage that goes beyond individual author inconvenience. When citations in medical papers are included to satisfy a reviewer rather than because they represent genuine scientific debt, the reference list stops functioning as a map of how knowledge actually built. A reader who follows that list to understand the evidence base for a clinical claim may encounter papers that are only there because someone with editorial power requested them. In clinical medicine, where treatment decisions can turn on what the literature says, that matters.

There is also a compounding effect over time. A highly cited medical paper attracts more submissions to the same journal or research group, which creates more opportunities for citation pressure, which further inflates the metrics that attracted the submissions in the first place. The system does not correct naturally once this feedback loop begins. It requires active intervention from publishers, funders, and the research community.

A 2026 commentary in PMC described this pattern as "editorial bullying," placing coercive citation within a broader category of practices through which editors and reviewers use their positional power to extract non-scientific concessions from authors. The authors of that commentary called for more transparent peer review models, independent oversight of editorial decision-making, and anonymous reporting mechanisms that authors can use without risk of retaliation.

Protecting Your Own Record

From the author's perspective, the practical goal is to build a citation practice you can defend. A reference list full of coerced citations is not yours in any meaningful sense. If the paper is ever scrutinized, those citations are part of the record. In fields like clinical pharmacology, oncology, and cardiology where post-publication audits have become more common, this matters.

Before submission, build your reference list from the literature rather than from anticipated reviewer preferences. This sounds obvious, but under deadline pressure there is sometimes a temptation to pad the list with highly cited or editorially important papers in the field to reduce friction. That practice is different from coercion but it shares a structural similarity: citations chosen for strategic rather than scientific reasons distort the record.

After receiving a review, evaluate citation requests on the same standard you would use for any other reviewer suggestion. Can you honestly describe this paper's relevance to your argument in a sentence? If yes, add it and describe why. If no, decline it and explain that the paper does not connect to your analysis. Either answer is professionally defensible. Neither requires aggression or lengthy justification.

Where This Stands in 2026

Citation coercion has been studied and condemned for over a decade, but the structural incentives that produce it have not changed. Impact factors still drive hiring and funding decisions in medicine and biomedical science. Journal editors who are active researchers still benefit from increased citations. The reviewer confidentiality that protects legitimate scientific critique also shields bad actors from accountability.

What has changed in 2026 is the monitoring infrastructure. Clarivate is more active in JCR enforcement. Publishers have added editorial board screening tools that flag unusual citation request patterns. Bibliometric transparency around citation networks is improving. Transparent peer review, which several major journals including BMJ and PLOS Medicine now offer, partially reduces the cover under which coercive requests can be made.

None of these changes make the practice disappear. But they do shift the calculus. Editors and reviewers who coerce citations are more likely to be caught and sanctioned than they were five years ago. Authors who push back politely and on record are in a better position than they were. The conversation around editorial integrity has matured enough that declining an unjustified citation request with a clear explanation is a realistic option, not a career risk.

The simplest practical takeaway: if you receive a citation request that you cannot explain in one sentence, do not add it. Document your reasoning, address the request in your revision letter, and treat it as you would any other unsupported reviewer claim. That approach is sound science, sound ethics, and sound self-protection.

Further Reading