Enforcing authorship standards
Guest authorship and ghostwriting are widely condemned and rarely sanctioned. Editors' bodies recommend sanctions and journals ask authors to document their contributions, yet journals, universities and licensing bodies each have reasons not to act, and conflicts over the author list are common, with independent dispute resolution proposed for them. Because a guest-authored article carries weight in court as well as in the literature, one legal argument proposes liability for the guest author who signs a false authorship warranty and for the company that uses the resulting article in court.
Editorial recommendations and journal requirements
The International Committee of Medical Journal Editors (ICMJE) has sought to curb inappropriate and unethical authorship practices by requiring journals to ask detailed questions about what each author contributed. The World Association of Medical Editors calls ghostwriting dishonest and unacceptable. It recommends that a journal that detects ghostwriting publish a notice that the manuscript was ghostwritten together with the names of the responsible companies and the submitting author, alert the authors' academic institutions and identify the commercial companies, supply specific names if contacted by the popular media or government organisations, and share the experience on the association's listserve and in other forums. The Committee on Publication Ethics recommends that editors adopt authorship or contributorship systems that make listings reflect who did the work and discourage ghost and guest authors, and that when the integrity of research is corrupted, errors and inaccurate or misleading statements be corrected promptly and with due prominence.
Some journals, including PLoS Medicine, have called for bans on future submissions by guests, formal retraction where unacknowledged ghostwriting emerges after publication, and reporting of misconduct to institutions. Such measures may affect academics concerned about their status and future publication options.
By 2010 the Council of Medical Specialty Societies, which represents about 650,000 of the estimated 920,000 physicians in the United States, had released a voluntary Code for Interactions with Companies; one of its provisions is that medical society journals should adopt policies prohibiting the submission of ghost-written manuscripts prepared by or on behalf of companies.
For several years after the ICMJE authorship criteria were introduced, scientists and physicians were frequently unaware of them. By 2010 journals frequently asked for documentation of authorship, and one medical writer judged that the situation was therefore likely improving. She warned, however, that authorship forms fall into the category of one more piece of paper to fill out, so writers would need to remind guest authors gently that the benefits of authorship come with the responsibility of explaining and defending the work after publication.
Most journals by 2016 required authors to meet at least the first three of the four ICMJE authorship criteria, and most companies required their employees to follow the Good Publication Practice guidelines, but that did not necessarily mean that ghost authorship had disappeared.
Why institutions do not act
Commentators have condemned industry ghostwriting with academic guest authors as unethical and unacceptable and recommended that journal submissions be policed more aggressively and that guest authors be sanctioned by journals, academic institutions and regulatory agencies. By 2013, however, the academic institutions, medical journals and medical licensing organisations that would seem to have the most at stake had not widely embraced these recommendations. Each of them has reasons not to act.
- Journals. It is unclear whether they can or even want to monitor the practice adequately, and some editors have said their journals are not responsible for policing authorship. Journals may also gain significant revenue from advertising and from selling reprints, including reprints of ghostwritten articles that industry may use for off-label promotion, so they are unlikely to act effectively against the practice.
- Universities. Several established academics have been associated with ghostwritten publications, but no public sanction appears to have followed. Several reasons explain the reluctance: the guest's role may be unclear, institutions may be challenged by their dual commitment to safeguarding academic integrity and protecting employees against unjust accusation, and universities tend to approach authorship questions with prudence given the serious potential effect on careers. Institutions may also be reluctant to act because ghostwriting cases often involve successful academics holding power through prestige, academic status, publication record and grant support. They may also decline to act because ghost and guest authorship may not be far removed from other common practices in academic medicine, where laboratory directors, departmental chairs and supervisors often claim authorship because of their roles rather than by standard authorship criteria. Some clinician-investigators even insist on co-authorship when they provide access to patients or samples. Pursuing sanctions for ghostwritten articles may lead to scrutiny of other authorship practices in academia, or to anxious efforts to justify them.
- Licensing and professional bodies. State medical boards in the US, colleges of physicians and surgeons in Canada and the General Medical Council in the UK could intervene when guest authorship by licensed professionals is uncovered, particularly where it involves outright misrepresentation of data. A physician who falsely claims to have analysed and adequately reported safety and effectiveness data can be considered to violate professional integrity standards and the commitment to patients and good health care; such a physician should know that this may impair clinical care and endanger patients, and should be sanctioned accordingly. These organisations have so far issued no serious sanction in the rare cases where one has looked into authorship allegations. The reasons may include general inertia towards new professional challenges and preoccupation with more traditional violations. These organisations have also been much criticised for a perceived tendency to protect the profession. For the same reasons as academic institutions, they may be uncomfortable confronting guest authorship and ghostwriting that damage their members.
Authorship disputes
The list of authors is a sensitive subject because researchers are assessed on their publications, and 40% of the individual files that the scientific integrity office of the French national institute of health and medical research processed over ten years concerned conflicts over it.
One analysis of authorship disputes concludes that researchers and journals should view the potential for disputes as predictable, preventable and soluble, rather than treating disputes as rare events to be handled case by case. It adds that independent bodies offering alternative dispute resolution services to scientific collaborators and to journals could quickly help research communities, particularly their most vulnerable members.
The value of an academic byline
Guest authorship violates the academic integrity standards that scientific reliability rests on. Sponsors borrow academics' names precisely because of the value and prestige attached to the presumed integrity and independence of academic researchers, and academics in turn receive considerable credit for publication, which gives them an incentive to act as guests.
In the legal setting, peer-reviewed articles are credible sources of evidence that may be used in lawsuits to support claims about safety and effectiveness, and hence to determine liability. Industry-controlled publications prepared by ghostwriters or using guest authors may distort perceptions of current knowledge about a product's safety and effectiveness. Under the standard the US Supreme Court set in Daubert v. Merrell Dow Pharmaceuticals, publication in a peer-reviewed journal is one of the criteria that help make a scientific theory or method admissible; by easing publication, guest authorship creates the impression that academic independence and integrity were satisfied when they were not, and makes it likelier that the research is treated as admissible when it should not be.
In the legal setting, the same articles also lend credibility to their guest authors. They are sometimes used to establish an expert witness's authority even where the validity of the research in the article is the very issue in dispute, so that treating the guest as a legal expert may prevent scrutiny of the practice being challenged for contributing to serious harm. Manipulations that influence the outcome of an industry-sponsored trial are often not visible to a guest author whose role in the study or article may be minimal and may fall short of the authorship criteria. The guest thus helps create the appearance that a study reflects the scientific methodology that Daubert requires for admissibility, and in the process is credentialed as an expert witness who can speak with authority about a product's efficacy and safety.
The authorship form as a warranty
One legal analysis argues that ghostwriting could be deterred more effectively by imposing legal liability on the guest authors who lend their names to ghostwritten articles. The argument rests on ghostwriting's bearing on the integrity of medical research and of scientific evidence used in legal disputes, and on the failure of journals, academic institutions and professional disciplinary bodies to enforce effective sanctions.
The legal argument starts from the authorship form. Medical journals typically require every author to confirm in writing that they meet the ICMJE criteria, which that analysis, in 2013, gives as three: substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; drafting the article or revising it critically for important intellectual content; and final approval of the version to be published, with all three required. Evidence revealed in class actions involving Vioxx, Prempro and Paxil suggests that guest authors often fail the first two: someone who reads an article or offers minor comments has contributed nothing substantial under either. The argument applies specifically to journals that require such a signed form as a condition of publication.
The authorship requirements are known to readers as well as to named authors, and the warranty is an important factor in judging an article's integrity and quality. An article prefaced by a statement that its lead author had refused to sign, or had repudiated, the authorship warranty and wished to clarify the contributions of an industry-based medical writer would have its credibility significantly undermined.
Fraud claims over a false warranty
Fraud occurs when a person knowingly makes a false representation in order to acquire something of value and harm results. In its basic structure, a civil fraud claim on this basis would take the same form in many countries, but such a claim is more likely to yield significant damages if many plaintiffs join in a class action, which may be done more easily in the US than in many other jurisdictions.
A guest's false warranty induces the journal to publish and misleads readers about the scrutiny the research received. The journal gives the guest credit for an article that may serve as a valuable credential with academic merit committees, grant agencies, conference organisers, and judges and juries if the guest later acts as an expert witness; such recognition may carry reputational and financial value. Arguably each repetition of the false warranty, implicit in a CV presented to any of those audiences, is an independent fraud. The journal loses the chance to publish an article that would have satisfied its requirements, and subscribers lose the chance to read one and may rely on data from a fraudulent article. A journal that knew the lead author was a mere guest would not publish the article.
The characterisation of guest authorship as fraud has received limited but important recognition in suits under the US False Claims Act, which imposes liability on those who cause fraudulent claims to be presented to the US government. In Strom ex rel. U.S. v. Scios, Inc. the government alleged conduct leading to false Medicare claims, including sponsoring ghostwritten articles purporting to validate off-label use of Natrecor (nesiritide) and, through press releases and sales representatives' promotion, recklessly encouraging prescribing for uses that were not medically accepted; without deciding the merits, the court held that the allegations if proved would state a claim. In Strom it appears to have been the unwarranted claims in the articles, rather than their fraudulent authorship, that supported the fraud allegation; the approach has great potential, but proving ghostwritten research false will not always be easy.
The fraud underlying these articles cannot be attributed solely to the guest, who after all has responded to an invitation. Pharmaceutical companies and medical communications agencies are well aware of the journals' publication requirements, and soliciting and facilitating the fraud may amount to conspiracy and incur liability on the same grounds as the fraud itself. Such conduct may also constitute fraud under the Racketeer Influenced and Corrupt Organizations Act, which applies to conspiracies involving at least two prohibited acts within ten years where those acts have the same or similar purposes, results, participants, victims or methods; the predicate acts include mail and wire fraud, which occur when a fraudulent statement is sent through the mail or by email. A guest who lends her name to two or more articles for the same product may satisfy those criteria in several ways, and civil liability under the statute allows treble damages.
Because all a journal's readers are harmed by the fraud, they may sue the guest in a class action under the statute: whether or not they read the article, its publication deprives them of an article meeting the journal's requirements and so diminishes the value of their subscription, which represents a year's worth of articles that conform to the guidelines. The harm may be measured by reducing the subscription price in proportion to the space devoted to the ghostwritten article. If a subscription costs $100 and the journal publishes 100 articles a year, each subscriber could be said to lose $1 per fraudulent article, an individually small loss that may be significant in aggregate, particularly if trebled under the statute. Pay-per-view purchasers, who make the same assumption about the article and would also be unlikely to pay after seeing a disclaimer of authorship responsibility, could form a distinct subclass with damages based on the download price. Plaintiffs would not have to prove individual reliance: the US Supreme Court held in 2008 that plaintiffs alleging fraud under the statute need not show they relied on the defendant's assertions provided they were harmed because someone else, such as the journal's editors, relied on the fraud. Establishing that the article was ghostwritten and that the plaintiff paid for a subscription or a download is then enough to establish fraud, reliance and harm for the whole class.
The approach aims at the guest rather than at the others complicit in the same fraud because claims under the statute added to negligence suits against pharmaceutical companies would yield low damages beside those already available. Against the guest, the combination of monetary sanction and reputational harm might deter academics, and might also deter the medical communications agencies that design the studies and seek impressive names for the byline. The threat of liability, and the uncertainty and distraction it brings, may be enough to discourage people who are not normally sued over harmful drugs but who help legitimate the studies that publicise them.
Fraud on the court
Against a pharmaceutical company that has used ghostwritten articles in litigation, the proposed route is also grounded in fraud: fraud on the court, a doctrine that takes a similar form in England, Australia, Canada, India and many other countries. Just as the integrity of medical research is key to recognising a false authorship warranty as fraud, the courts' concern for the integrity of their proceedings is key to this doctrine. A recent US formulation defines it as conduct by an officer of the court, directed at the judicial machinery itself, intentionally false or wilfully blind or in reckless disregard for the truth, consisting of a positive averment or a concealment where there is a duty to disclose, that deceives the court. That definition would apply to ghostwritten articles cited by the lawyers, or the expert witnesses, of the parties that helped create them. Expert testimony reaches a court through lawyers, who are officers of the court; where a company helped produce ghostwritten articles and its lawyers cite them, the lawyers are at the very least reckless about the falsehood and have a duty to disclose the truth. Remedies may include default judgment for the opposing party when the fraud is revealed during a proceeding, nullification of a judgment or legal entitlement secured with the fraud's aid, and disbarment of the counsel who facilitated it.
The doctrine began with a ghostwriting case, and Hazel-Atlas Glass v. Hartford-Empire (1944) seems to be the only ghostwriting case the US Supreme Court has decided. In 1926 Hartford-Empire tried to patent a method of moulding glass. Faced with a sceptical Patent Office, Hartford's employees wrote an article praising the method and found an author for it in the president of a glass workers' union; the article was published in a trade journal and cited in the patent application, and the patent was granted. Hartford sued Hazel-Atlas, a competing glass manufacturer, for infringement and lost at trial. On appeal Hartford leaned heavily on the article, and Hazel-Atlas, doubting it, interviewed the nominal author, who refused to acknowledge the truth. The court of appeals ruled for Hartford, quoting the article as evidence of the patent's novelty and utility. The truth emerged nine years later when Hartford disclosed its files in an antitrust action. In 1944 the Supreme Court vacated the judgment as a fraud on the court and nullified the patent, and the lawyers who had used the article were barred from practice before the Patent Office.
Observations from that judgment apply with equal force to medical ghostwriting. Using spurious claims of authorship to legitimate claims before the Patent Office and the courts is a wrong against the institutions set up to protect and safeguard the public, and the same can be said of ghostwritten articles published in medical journals through false warranties of authorship, with the courts among the institutions wronged, since judges may treat the resulting publications as admissible under Daubert. And Hartford's argument that the article could not be shown to have caused its victory was rejected: its officials and lawyers thought the article material, went to considerable trouble and expense to get it published, urged it on the court and prevailed, and were in no position afterwards to dispute its effectiveness. The same answer would be appropriate to a pharmaceutical defendant minimising the evidentiary role of a ghostwritten article today.
For a sponsor, the risk of sanction could restrict the legal use of its ghostwritten articles, not their use in promotion. Ghostwritten articles are not created primarily for legal purposes but to publicise and market drugs, yet they are often used in litigation to support the manufacturer's arguments about efficacy and safety, to establish scientific acceptance for Daubert purposes, and to credential expert witnesses. If each of those uses risks sanction when attempted by a party that helped create the article, its overall value could fall significantly: the article can still promote the drug, but if litigation arises the defendant's arsenal is limited.
Accuracy as a defence
Whether a publication with an academic guest author is factually accurate is irrelevant to the objection. In Hazel-Atlas Hartford insisted the article's claims were true notwithstanding the attribution, and the Supreme Court found the argument unavailing: truth needs no disguise, and the article, even if true, should have stood or fallen under the only title it could honestly have been given, that of a brief on behalf of Hartford, prepared by Hartford's agents, attorneys and collaborators. Sponsors of ghostwritten articles may be expected to treat false authorship as an insignificant detail meriting no legal sanction, and the Court's comments are a sufficient rebuttal. The false respectability that academic investigators lend to claims of safety and effectiveness risks undermining the integrity of biomedical research and patient care, and the same integrity underpins the use of scientific evidence in a courtroom.