In health sciences, inappropriate authorship (IA) is one of the commonest questionable research practices and could be equally or more pernicious than scientific fraud.123 Here, I focus on coercion authorship (CA) as usually enforced by principal investigators or group leaders and its detrimental effects on the whole scientific endeavor456 but mainly on graduate students and young researchers who may suffer from frustration and leave their academic careers.7 Excluding some historical authorship abuses by mentors,7 CA has been documented since 1996 and recognized around the world, but mostly fought against in wealthy nations.8910 The spreading and severity of CA in low- and middle-income countries (LMICs)11 are exposed in the stinging remarks by authors from those regions.1213

Despite the pervasiveness of CA, its occurrence and major determinants have been mainly explored in surveys to graduate students and junior scientists in Scandinavia. Since 2013, Bjørn Hofmann’s team has gauged the pressure exerted on doctoral students (mainly in Medicine) to include free-riding authors in or modify the byline. In 2023, Hofmann et al.14 reported that 12.8% of doctoral students (n = 536 respondents to anonymous surveys) from the Medicine Faculty of the University of Oslo in the 2010-2020 period declared to have undergone CA; moreover, among those who got their doctoral degree in 2019 (n = 94 respondents), the mean was 26.6% and increased to 40.4% when students whose answer was “uncertain”, were added.

Also in Norway, another team15 asked online ≈2,700 faculty, researchers, and PhD students at the Medical Faculty and Hospital from the University of Oslo on authorship issues. Of the 654 respondents, 233 (36%) “had experienced pressure to include undeserving (i.e. ‘guest’ or ‘honorary’) authors in their articles”.

In another anonymous survey,16 the authors analyzed the responses of 1,336 doctoral students in 5 disciplinary areas or faculties (law, STEM, medicine, social sciences, and humanities) from universities of Danmark, Hungary, Ireland, Portugal, and the francophone region of Switzerland (n = 202-427 students per country). Globally, 377/1,336 (28%) students recognized to have granted or consented at least 1 CA to “the person in power”, ratio that increased to 34% among the 1,096 students who had already published 1 or more papers. The percentages per faculty varied from 10 (law) to 49 (medical sciences) and per country from 19 (Hungary) to 43 (Portugal). The most common justification (49%) given by students who underwent CA was that the powerful person plainly requested to be included as a coauthor. Of note, an independent study17 based on semi-structured interviews with 33 biomedical researchers (young, mid-level, and seniors) from 5 Swiss universities, reported on the respondents’ reactions to a hypothetical insistence of the department head to be granted authorship without deserving it. Even if most researchers found this claim morally unacceptable, the authors concluded that “junior researchers will continue to be challenged with the dilemma of how to accommodate the uninvited guest author”.

CA has also been documented elsewhere. I refer here to 3 online confidential surveys. In 1996, 193 first authors of major articles published in a radiology journal during 1992-1993 provided information on the respective bylines. Out of around 920 authors, there were 17% honorary authors and, among these, 12% demanded directly or indirectly their inclusion as coauthors.8 In another study, Canadian scientists surveyed the corresponding authors (n = 127 respondents) of original articles published in 4 top medical journals; the results showed that 16% of the participants had experienced CA.18 In Croatia, Ljubenković et al.19 applied Hofmann’s questionnaire to 1,177 individuals affiliated to the Zagreb School of Medicine. Apart from college and graduate students, the sample included PhD candidates (n = 265), recently graduated doctors in medical sciences (n = 95), and mentors or supervisors (n = 370). Though few individuals answered the survey in the last 3 categories (41, 14 and 31, respectively), the ratio of CA they recognized was 24.7, 35.7, and 41.9% while the perceived risk of suffering severe consequences by incurring in authorship malpractice was 12.2, 21.4 and 6.5%, respectively.

Even if no surveys focusing on CA in LMICs are available, a recent overview of research integrity and misconduct alludes to gift or guest authorship and refers to a few studies on IA in those countries.20 To further illustrate this point, I remark that in Mexico researchers are aware of both the rampant occurrence of authorship transgressions such as CA212223 and the lack of institutional plans for promoting integrity, focused studies, and effective deterrents or sanctions addressed to curb CA and other authorship malpractices. As an analysis of retracted articles from Kazakhstan reveals,24 CA is not a recognized cause of retractions (probably because it occurs before publication) or is included in the general term “authorship issues”. In this sense, the “publish or perish” mindset enhances the formidable challenges that editors in the developing world face since they must work despite poor publishing infrastructure and ethical regulations.25 Likewise, editorial efforts in emerging economy countries such as South Korea,26 India,27 and Portugal28 target IA. Furthermore, the “guanxi” culture so ingrained in China explains how authorship order often reflects the hierarchical structure in which doctoral students are expected to concede the first authorship to their mentors.29 Admittedly, this may allow CA and guest authorship to be subtly disguised as honorable practices.

Developing countries must promote integrity and responsible research practices while establishing strategies to restrain misdemeanors. Below, I highlight some actions advanced mainly in developed countries but also feasible in LMICs:

1. To set up formal college and graduate instruction on integrity, publishing ethics, and responsible authorship.13 These courses should focus not only on authorship criteria and common malpractices but also on the availability of tools, institutional procedures, and online resources that help students to face conflicts and dilemmas.16 Thus, students could demand transparency in authorship and protect themselves against arbitrary actions of senior researchers and mentors.3031

2. Professors and mentors should also receive the same or similar instruction to reinforce their role of models and to assist them in building a work environment where ethical practices prevail and complement the theoretical lessons.14 In addition, researchers should continuously and openly discuss ethical issues including authorship concerns with colleagues and students.20

3. Universities and research centers must publish their own authorship norms and policies.13 In addition, these institutions should establish a system to deal with authorship abuses or disputes31 and even appoint an ombudsperson32 to support graduate students and junior scientists as well as to monitor the compliance of the aforementioned policies. Recent initiatives in Brazil to promote best practices and appoint an ombudsperson are noticeable.33

4. The continuous growth of inter-disciplinary studies authored by dozens or hundreds of scholars, including graduate students, led Smith and Master30 to advance a 5-step collegial schema to fairly assign authorship or contributorship credit. A comparable 2024 roadmap for addressing faculty-student authorship issues is also worth reading.7

Everywhere, the primary responsibility of senior and principal researchers to promote fair authorship practices and discourage unfair ones is even greater when it comes to CA due to the asymmetrical power relationship between experienced authors and novice scientists. In LMICs, research institutions have a strong need to foster scientific integrity and therefore should publish their own authorship policies, set up the proper mechanisms to reinforce ethical authorship, and supervise the compliance of the published norms. Although CA in submitted manuscripts can hardly be spotted by editors and reviewers, these gatekeepers are in a good position to detect guest authors43435 and thus indirectly prevent some cases of hidden CA.