Abstract
Background
This study aims to determine the prevalence of perceived unfairness in authorship inclusion and byline ordering, two common concerns in the literature on authorship, among a sample of the physician population at the University of North Carolina School of Medicine.
Methods
We developed and distributed a survey among eight different departments at a large U.S. academic medical center, with questions pertaining to basic demographics, contributions to research, and perceptions of authorship inclusion and ordering on their publications. Responses were analyzed using chi-squared test for predictors of perceived author order misplacement or exclusion from authorship.
Results
Out of 83 respondents (9.2% response rate) from eight clinical departments, 52% perceived unfair authorship ordering and 34% perceived unwarranted exclusion from a project. Perceived lower positioning of the author byline was reported to occur more commonly in the earlier stages of training (medical student and resident levels). The most common reasons for these occurrences were related to issues with communication.
Conclusions
Difficulties around communication and the hierarchical nature of medical training contribute to the perceived unfairness of author ordering and authorship inclusion. Structured authorship agreements and education in publication ethics may reduce these perceived inequities.
Introduction/Background
Authorship has long been described as the currency of the research world,1,2 driving the trajectory of a researcher's career in terms of establishing tenure, obtaining grants, and enhancing one's overall prestige. 3 First studied in 1967, research on authorship experienced an upward trend in the 1990s, especially in the health sciences. Two central concerns remain consistent in the literature: the inclusion of authors and the order in which they appear. 1
While several guidelines have emerged to address these issues, such as the ICMJE authorship criteria and CRediT taxonomy, they are not without debate.4–6 For instance, as found by Maggio et al, these guidelines do not provide instruction in areas of ambiguity, such as authors whose contributions change over time due to uncontrollable events (eg illness), and thus are open for individual interpretation in these circumstances. 7 Furthermore, respondents in both Maggio et al and Mavis et al reported instances where authorship criteria could be bent for seemingly benevolent reasons, such as assisting more junior faculty in advancing their career.7,8
This lack of clarity in research authorship can be a challenge for those who mentor medical trainees, and medical trainees may have widely differing experiences in whether they learn about proper authorship practices and how they are taught to implement these guidelines. In a survey study by Karani et al, they found that 26% of medical students involved in research did not have discussions around authorship and 66% of medical students reported not having had formal education on authorship guidelines. 9 Rajasekaran et al found that about 90% of postgraduate medical trainees were unaware of the ICJME authorship guidelines. 10 This may subsequently lead to disputes in authorship in later stages of training and beyond.
To explore these issues around authorship, we sought to determine the prevalence and predictors of perceived authorship unfairness and explore potential contributing factors. We chose a population across a broad level of training to better study issues that may be more specific to those still in medical training and to identify areas in which these problems could be addressed at the level of both the individual and the institution.
Methods
In this cross-sectional study, we developed a 36-item survey centered around themes raised in the literature on authorship (Supplemental Material 1), as we did not identify an existing validated instrument that directly matched the specific aims of this study. This survey was tested with an initial pilot group of physician researchers in academia to assess for areas of refinement in addition to technical usability; the results from this pilot study were not used. The final product was revised based on this feedback and included four sections with questions about respondents’ basic demographics, contributions to research, and perceptions of authorship inclusion and ordering on their publications. The questions on basic demographics included biological sex, race/ethnicity, medical specialty, and level of training (resident, fellow, or attending physician). Respondents were also asked about their participation in research and the capacities in which they contributed. Finally, they were asked about authorship ordering and inclusion on these projects: specifically, whether they perceived themselves to be placed higher or lower than what they felt they deserved and if they had ever perceived unwarranted exclusion from authorship.
The survey was distributed via a Google Forms link to all resident, fellow, or attending physicians of the following departments at the University of North Carolina School of Medicine which had agreed to take part in the study: General Surgery, Internal Medicine, Neurology, Obstetrics and Gynecology, Ophthalmology, Orthopedic Surgery, Urology, and Otolaryngology. Residents, fellows, and attending physicians were eligible to participate. Participants were given 2 weeks to complete the survey, with two reminder emails sent within this time. Involvement was entirely voluntary, with no compensation for participating. Respondents were only allowed one submission, and responses were anonymous. Incomplete surveys were not included in the analysis. The reporting of this study conforms to the CROSS guidelines 11 (Supplemental Material 2). Data collection took place from July 18 to July 31, with a reminder email sent on July 25.
The University of North Carolina Office of Human Research Ethics determined this study to be exempt from further IRB review (Study #21-1873; Exemption Category 2 – survey procedures; 45 CFR 46.104; determination date August 13, 2021). Participation was voluntary and responses were anonymous. The survey introduction included a participant information sheet; written/signed informed consent was not obtained, and consent was implied by completion and submission of the survey.
Statistical Analysis
Chi-squared analysis was performed for predictors, including status of attending physician and sex, of perceived author order misplacement or exclusion from authorship using Microsoft Excel (v16 l.56, 2021). Only complete survey submissions with responses to all the questions were included in the analysis. We considered p < .05 as statistically significant.
Results
The survey was sent to 902 individuals and received 83 respondents, yielding a 9.2% response rate. Of the 83 respondents, 51.8% were male, with 63.8% identifying as attending physicians. More than half of the respondents felt they had been credited in a lower authorship position than deserved (51.8%), and 33.7% reported being excluded from a publication in which they contributed. Notably, lack of formal discussions was the most frequently cited cause for both lower authorship placement and exclusion. Approximately 25% of participants felt they had been credited in a higher position than deserved.
Respondent Demographics and Research Contributions
Eighty-three participants responded to the survey with their demographic characteristics, including sex, race/ethnicity, and medical specialty (Table 1). Thirty of the respondents were in residency or fellowship and 53 were attending physicians, with a wide range of years in practice (1–47 years, mean 12.6 ± 10.6) (Table 2).
Characteristics of Respondents.
Seniority Breakdown of Respondents Who Identified as Either Residents or Fellows.
Notes. PGY: Post-Graduate year.
All the participants reported having contributed to a research study and specified the specific roles in which they contributed to their studies. The average number of publications was 44.6 ± 60.8 publications with a range of 1–300 publications (Table 3).
Scientific Contributions of Respondents.
indicates respondents were able to select more than one option.
Perceptions of Authorship Fairness
When asked about the perceived fairness of their scholarly authorship, 34 (41%) of respondents felt they were placed appropriately on the author byline in all their published papers. On a follow-up question, 69.9% of respondents reported this perceived fair placement occurring in residency and 45.8% reported this occurring in fellowship (Table 4). More than half of respondents (51.8%) reported feeling that they had been credited in a lower authorship position than they believed was warranted. These respondents felt this occurred more often as a medical student or resident (39.8%) than as an attending physician (16.9%). The top reasons for having been placed in a lower authorship position were having had no formal discussion of authorship order (27 [32.5%]), the PI or mentor positioning them out of a higher authorship position on the project (19 [22.9%]), or miscommunication (9 [10.8%]) (Table 5).
Respondents’ Experience with Fair Authorship Placement.
indicates respondents were able to select more than one option.
Respondents’ Experience with Higher and Lower Authorship Placement.
indicates respondents were able to select more than one option.
A small group of respondents [21 (25.3%)] felt they had been placed in a higher position than they deserved (Table 5). There was no significant difference between the group that was in training (16.9%) and the group composed of attending physicians (15.7%).
Authorship Inclusion and Exclusion Experiences
We also examined respondents’ experiences of being included or excluded from authorship without prior consent (Table 6). Three (3.6%) respondents reported being credited without their knowledge, all occurring during residency. More respondents reported being excluded from a publication than being credited without consent [28 (33.7%) versus 3 (3.6%)]. Most of the respondents who reported exclusion stated this happened as a medical student or an attending physician (11 [13.3%] for each); this occurred less often as a resident or fellow (7 [8.4%]). The reasons for exclusion were lack of formal discussion for authorship order (18 [21.7%]), having been perceived as not having contributed enough (10 [12%]), and miscommunication (7 [8.4%]). When 16 (19.3%) of respondents questioned or challenged their assigned authorship, 7 (43.8%) of them felt this allowed them to change their authorship (Table 6).
Respondents’ Experience with Inclusion and Exclusion in Authorship.
indicates respondents were able to select more than one option.
We found no significant association between attending status or sex and being placed in a higher or lower position in author order or being excluded from authorship (P > .05) (Table 7). Due to the small number of responses per category, it was not possible to analyze the results according to author ethnicity or specialty.
Chi-Square Analysis on the Relationship of Attending Status and Sex to Perceived Improper Authorship Ordering and Authorship Exclusion.
Discussion
Our results showed more than half of respondents reporting having been placed in a lower position on the authorship byline than was felt to be deserved. Additionally, nearly a third of respondents felt they had been excluded from authorship on projects they contributed to. This is similar to the findings in a study by Uijtdehaage et al, where they found that about 35% of authors in the health professions education community experienced exclusion from authorship. 12 Both issues (author ordering and authorship inclusion/exclusion) have been previously studied and may worsen due to the growing number of collaborative studies and authors per paper in recent years. 1 This can lead to an increase in research misconduct and negatively impact the research community. 13 Similarly, as found in the present investigation, perceived unfairness may increase.
Our findings suggest that authorship disputes are particularly prevalent during the earlier stages of training, possibly due to power dynamics between trainees and senior investigators. 14 The perceived power imbalance between a student and professor/supervisor has been shown to contribute to authorship disagreements that are often left unresolved. 1 Our research offers insights across different training levels, revealing the hierarchical nature of authorship disputes as early as medical residency. In addition to the power imbalance, these disputes can also be due to the transient nature of the trainee role relative to the more permanent role of the senior investigator. For example, once a medical student or resident graduates from their program, it can be more difficult to stay in contact with the research team, and if a project is particularly drawn out, the senior investigator may also forget who was involved and their specific contributions to the team. Both the power imbalance and higher turnover of medical trainees can lead to having more authorship disputes in the earlier stages of training.
Although uncommon in this study, being included as an author without one's knowledge was experienced at the stages of resident, fellowship, and attending physician. Fewer inconsistencies with authorship inclusion/exclusion were perceived during fellowships. This could be because fellowships are optional, shorter, and sometimes require primary authorship.
We also investigated the reasons for perceived lower position in the author byline or authorship exclusion. Communication-related difficulties were the top reasons for perceived lower positioning in author order; “no formal discussion” and “miscommunication” were chosen by approximately two-thirds and one-fourth of respondents, respectively. The second-most cited reason for perceived lower author position was being placed lower by more senior members (22.9%). The hierarchical nature of medical training and the medical field could contribute to difficulties with open and honest conversation, especially when trainees need to initiate discussions with principal investigators or when lower-ranking assistant professors must communicate with higher-ranking professors or departmental leaders. In our survey, at least half of the respondents felt they had been assigned to a lower position on the byline or experienced an underserved exclusion from a publication. Within this group, less than half reported questioning or challenging these decisions, with half of those individuals successfully obtaining the results they felt matched their level of contribution. The small number of individuals who attempted to change the authorship decision could point to the difficulty associated with having these discussions. Other reasons for having a lower position on the author byline that are related to the decision-making of the senior author could be gift authorship, in which individuals who contributed little are named as authors, or sponsorship, in which senior researchers attempt to help their juniors advance in their careers by excluding or deemphasizing the contributions of other authors or falsely magnifying the contributions of their own students (such as by placing them in the first author position). 3
On the other end of the spectrum, 25% of respondents reported being placed higher in the author byline than they expected. This may stem from discrepancies in perceived effort or variations in the senior authors’ assessments of contributions, as well as instances of gift authorship or sponsorship.
For perceived authorship exclusion, dilemmas around communication were again the top reason among those surveyed, with “no formal discussion” and “miscommunication” chosen by approximately two-thirds and one-fourth of respondents, respectively. The second-most selected reason was the perception that the respondent didn't contribute enough to the project. As our results showed, this occurred throughout all stages of medical training, this shows the prevalence of the power dynamic between the supervisor or most senior author and the rest of the authors in making decisions on authorship. Thus, many of the same factors explaining lower authorship position also apply to authorship exclusion.
Our analysis revealed no significant influence from respondents’ sex or training status on perceived unfairness in authorship ordering or exclusion. The lack of significant differences based on sex or training status suggests that authorship disputes are a universal challenge within academic medicine, transcending traditional demographic boundaries, or are merely artifacts of the study design. Regardless, this finding points to systemic issues in research culture that affect all contributors, regardless of their seniority or sex.
As previously mentioned, authorship in academic research has long been a point of contention and study. The consequences of unethical authorship are far-reaching, ranging from distrust between the general public and the scientific world to potential patient harm.13,15 Despite the increase in research around this topic, successful resolution of authorship dilemmas continues to remain elusive. This could be due to the entrenchment of old practices and lack of clear guidelines. First, the hierarchical nature present in academic institutions can make it difficult for trainees and junior faculty to speak up or against unscrupulous practices.1,15,16 Furthermore, unethical authorship practices may actually benefit other parties in addition to the author committing the unjust deed. As mentioned by Aliukonis et al, with honorary authorship practices such as gift authorship, the primary author also stands to gain more publications to their name and the academic institution may also receive more prestige and chances for funding, contributing to a cycle based on academic capitalism. 15 Second, while some journals provide authorship criteria, others either provide guidelines that are vague or do not have guidelines at all. 15 Issues have also been raised with the guidelines currently in use. The second criteria proposed by the ICMJE, which states “drafting the work or reviewing it critically for important intellectual content”, has been critiqued as unrealistic and restrictive, especially in multidisciplinary research. 15 The ICMJE criteria also does not provide guidance on determination of authorship order. 5 Given the deeply rooted current practices and the absence of a clearly agreed upon set of guidelines to follow, it is not surprising that elimination of unethical authorship seems to be at an impasse.
However, promising avenues for approaching this dilemma have been proposed, three of which include having support systems for resolving authorship disputes, promoting early education of publication ethics, and introducing concrete methods for determining authorship. First, regarding the creation of support systems at academic institutions; these would not only serve as neutral third parties in mediating such disputes but also as a way to expose junior members, such as students and junior faculty, to the nuances of publication ethics and also promote discussion around authorship in general.15,17 Second, introducing publication ethics education early in training, such as during the undergraduate years or in medical school, can help inform trainees and provide them with a basic understanding of fair versus unfair practices. 15 This may help decrease the chances of authorship disputes occurring due to more knowledge and awareness of them at an early stage of instruction. Finally, a variety of methods for more objective determination of authorship have been proposed, such as the CalculAuthor, authorship grids, and a points-based author score.18–20 These methods, based around defining authorship order and inclusion by delineating specific contributions and responsibilities, offer the advantages of establishing communication from the beginning of the project, keeping all parties abreast of their own duties and that of others, and providing a tangible log that can be used both among the members of the research team and journals as clear evidence of author contributions and subsequent inclusion and byline ordering. While the actual implementation of these structured authorship agreements may differ depending on the institution or other contexts, practical implementation could be done at a higher level, such as academic centers providing education on methods that have already been proposed (like the aforementioned authorship grids or points-based system), and at the individual level, such as research teams collectively choosing a system to use at the start of a project and establishing key milestones at which to revisit it. These three solutions – creating support systems for resolving authorship disputes, introducing early education of publication ethics, and implementing formal methods for determining authorship – can be utilized by training programs to teach proper research conduct and handling of authorship conflicts.
Our study has limitations, including its single-institution focus, a low response rate from trainees, and a relatively small overall response number, which constrains the generalizability of the findings. Additionally, the low response rate raises the potential for nonresponse bias, which also affects the degree to which the data represents the general population. This was an exploratory survey, and no sample size calculation was performed prior, which could result in this study being underpowered. Furthermore, formal psychometric validation and reliability testing were not performed. Not all specialties at our academic institution were sampled. The idea was conceived among discussion with physician leaders from multiple departments resulting in a de facto cohort of medical and surgical specialties at our institution. Medical students were not included in this study; their responses may have revealed more instances of perceived unfairness compared to the other groups of respondents (residents/fellows/attending physicians) and could have placed more weight on the influence of the hierarchical nature of medical training on perceived unfairness in authorship exclusion and ordering. It is important to note that this study's findings are based solely on respondent perceptions and are thus potentially influenced by recall bias and subjectiveness of responses. However, these data still offer valuable insights into miscommunication or disagreements since regardless of validity; they are still accurate representations of perceived fairness. Additionally, our study, conducted at the University of North Carolina, includes faculty with diverse training and work backgrounds, making it representative of the academic physician workforce. 21 Future studies could address these limitations by including a larger, multi-institutional sample and incorporating medical students in addition to resident, fellow, or attending physicians to capture a broader range of experiences. Additionally, qualitative interviews could provide deeper insights into the interpersonal dynamics that contribute to authorship disputes. Finally, testing the survey's validity and reliability will be a key direction for future work.
While disputes around authorship may not truly ever disappear, we remain hopeful that the continued awareness of such conflicts through studies such as this, as well as the introduction of solutions from others’ works, will drive an overall positive change to this area of publication ethics.
Conclusion
Perceived unfairness in authorship remains prevalent across training levels. Addressing communication gaps and hierarchical barriers through structured authorship discussions and ethics education may improve fairness in collaborative research.
Supplemental Material
sj-pdf-1-mde-10.1177_23821205261431001 - Supplemental material for Perceptions of Academic Authorship in Medicine: A Multi-Specialty Survey of Physicians
Supplemental material, sj-pdf-1-mde-10.1177_23821205261431001 for Perceptions of Academic Authorship in Medicine: A Multi-Specialty Survey of Physicians by Lauren Dimalanta, Dipen Kumar, Alice Yang Zhang, Shannelle Campbell, Maureen Dale, David Friedlander, Ganesh Kamath, Lauren Schiff, Brian Thorp and David Fleischman in Journal of Medical Education and Curricular Development
Supplemental Material
sj-docx-2-mde-10.1177_23821205261431001 - Supplemental material for Perceptions of Academic Authorship in Medicine: A Multi-Specialty Survey of Physicians
Supplemental material, sj-docx-2-mde-10.1177_23821205261431001 for Perceptions of Academic Authorship in Medicine: A Multi-Specialty Survey of Physicians by Lauren Dimalanta, Dipen Kumar, Alice Yang Zhang, Shannelle Campbell, Maureen Dale, David Friedlander, Ganesh Kamath, Lauren Schiff, Brian Thorp and David Fleischman in Journal of Medical Education and Curricular Development
Footnotes
Acknowledgements
We would like to thank the study participants for their time in completing the survey.
Ethics Approval
The University of North Carolina Office of Human Research Ethics determined this study to be exempt from further review (Study #21-1873; Exemption Category 2 – survey procedures; 45 CFR 46.104; determination date August 13, 2021).
Consent
Written informed consent was not obtained. Participants reviewed a study information sheet on the survey landing page, and consent was implied by voluntary completion and submission of the anonymous survey.
Consent for Publication
Not applicable; the manuscript reports only aggregate, de-identified data.
Authors’ Contribution
LD: writing – original draft, writing – review and editing, visualization; DK: writing – original draft, writing – review and editing, data curation, formal analysis; AYZ: writing – review and editing, supervision; SC: writing – review and editing, supervision; MD: writing – review and editing, supervision; DF: writing – review and editing, supervision; GK: writing – review and editing, supervision; LS: writing – review and editing, supervision; BT: writing – review and editing, supervision; DF – conceptualization, writing – review and editing, supervision.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
