Research integrity and academic medicine: the pressure to publish and research misconduct

Authorship

The number of publications continues to increase, nearly doubling since 2010 [2]. Over the past 30 years, the average number of authors listed in an article has also increased, creating an opportunity for misrepresentation [3]. Being listed as an author comes with benefits such as prestige, increased funding, and career opportunities [3]. The International Committee of Medical Journal Editors (ICMJE) is a group that develops recommended protocols for conducting scholarly work [10]. ICMJE has suggested four criteria that should be met in order to be considered and named as an author on a scientific work:

1. Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work

2. Drafting the work or revising it critically for important intellectual content

3. Final approval of the version to be published

4. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved [10].

The fourth criterion was added in 2013 to put an emphasis on accountability for all authors listed on the paper [7]. Adhering to these authorship criteria may prevent misrepresentations and hyper-authorship. Many well-known peer-reviewed journals follow the ICMJE suggestions; however, individual journals can set their own criteria for what constitutes an author [11].

There are multiple checkpoints that a project goes through to preserve scientific integrity and prevent research misconduct. Fostering a culture of academic honesty begins at the local institutional level with strict regulatory oversight by federal regulatory bodies, such as the US Office of Research Integrity (ORI) [12, 13]. The three major safeguards in place to maintain scientific integrity are: peer review for funding; the referee system of peer review for publication; and replication of results [13]. When these checkpoints fail, the possibility for misconduct to occur increases.

Research misconduct

Research misconduct is defined by the US ORI as “the fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting results” [12, 14, 15]. There are numerous types of infractions in integrity with a range of seriousness. Plagiarism may present as superficial paraphrasing or the use of exact words without proper citation, or it may come in the form of utilizing others’ ideas and misrepresenting them as one’s own [2, 16]. During the process of data collection, research misconduct may take place in the form of data “trimming,” which is the removal of irregular results to strengthen desired results, or data “cooking,” which involves the deliberate manipulation of data to produce the desired results [2, 12]. The most egregious form of misconduct is the overt fabrication or falsification of data or results. Additional definitions related to misconduct are listed in Table 2.

Although not strictly covered by the ORI’s definition of research misconduct, there are a number of improper research practices that should be noted. One of the most common improper practices is assigning “honorary authorship,” which involves listing undeserving authors on a publication [2, 12, 17]. The motivations for this practice can be vast and variable. For example, honorary authorship may be offered by a junior researcher as reciprocity to a senior researcher who holds rank, influence, or funding in the department. Conversely, a senior, influential researcher may coerce a junior researcher to list them as an author without having met the criteria of authorship as defined by the ICMJE [2, 10, 17]. Alternatively, a junior researcher may list a highly influential honorary author as a tool to increase the likelihood of obtaining publication or funding from an entity that would otherwise be unattainable [2, 17]. With regard to misconduct during the peer review process, an influential author could generate bias, either positive or negative, from the reviewer [2].

Scholarly activity in medical education

For medical students, the motivation or pressure to engage in scholarly work may be multifactorial. To maintain accreditation, both allopathic and osteopathic medical schools in the United States must demonstrate support for faculty- and student-driven research. They must produce evidence of student participation in research and scholarly activity [18, 19]. Students may feel institutional pressure to participate in research as the medical school aims to maintain accreditation.

In addition to completing their academic work and clinical training, medical students must actively prepare to apply to postgraduate training programs, commonly known as “residency programs.” Securing and successfully completing residency training is imperative to the unrestricted practice of medicine in the United States. Students apply for residency during their final year of medical school and must undergo a rigorous application and interview process to match into a program. Residency programs rank applicants based on numerous characteristics such as academic performance, including United States Medical Licensing Examination (USMLE) and Comprehensive Osteopathic Medical Licensing Examination (COMLEX) scores.

Furthermore, residency programs evaluate students based on participation in research and scholarly activity, service, leadership, and interpersonal skills. Prior to 2022, USMLE Step 1 and COMLEX Level 1 were graded numerically. Both USMLE Step 1 and COMLEX Level 1 transitioned to Pass/Fail scoring in 2022, which has potentially far-reaching implications for medical students and residency programs. It is speculated that the loss of numeric board scores will prompt programs to weigh other elements of the application more heavily, including research and scholarly activity. In fact, results from the 2022 residency match reveal that more research experience and publications are associated with higher residency match rates for medical students in all specialties for allopathic students and most specialties for osteopathic medical students [20, 21]. For students pursuing surgical specialties, such as general surgery [22], orthopedic surgery [23], or neurosurgery [6], research productivity plays a substantial role in matching into residency.

Facing pressures similar to medical students, resident physicians and those in fellowships frequently find themselves entrenched in a “publish or perish” culture [3]. In the United States, residency training programs are overseen by the Accreditation Council for Graduate Medical Education (ACGME). For the program to maintain accreditation and for individual residents to remain in good standing, participation in scholarly activity is required by the ACGME [24]. Furthermore, research and other scholarly activity encompass an important part of applications for resident physicians seeking fellowship programs. Unfortunately, evidence suggests that a substantial minority of students and resident physicians misrepresent their scholarly work [8, 25], [26], [27].

In addition to misconduct during the research and publication process, there exists potential for misconduct and misrepresentation in reporting one’s work. It is not uncommon for job applications, residency or fellowship applications, and curricula vitae (CV) to include misrepresentations of varying severity [8, 25], [26], [27]. From egregious actions such as listing fake publications to lesser infractions such as embellishing the status of the listed work, misrepresentations are not rare [8, 25], [26], [27]. Yeh et al. [25] found that one in eight (12 %) student candidates interviewing for general surgery residency at their institution had serious inaccuracies in reported vs. actual publications. Oke and Mantagos [28] found that 4.7 % of applicants to a pediatric ophthalmology fellowship listed unverifiable publications on their applications. Cortez et al. [8] examined the accuracy of publications listed by applicants to an orthopedic sports medicine fellowship program. Within their cohort, 68 % of the articles were reported as “completed,” with 5.7 % of those being found to be inaccurate, and 31.6 % of the articles were reported as “submitted,” with 28.3 % being unpublished between 2 and 6 years later [8]. In order from most to least common, the types of misrepresentation found by Wiggins [26] was listing nonexistent articles, authorship order, and nonauthorship. This analysis also revealed that there was no consensus as to what constitutes misrepresentation [26]. None of these studies discuss if there were any consequences faced by the applicants, but they do reveal that misrepresentation has been a problem for many years.

Scholarly activity by faculty physicians

Physicians who pursue a career in academic medicine, commonly referred to as “faculty physicians,” encounter numerous motivations to produce scholarly work and publications. As with medical students and resident physicians, faculty physicians face regulatory mandates from accrediting bodies to engage in research and scholarly activity [18, 19, 24]. Additionally, institutional pressure frequently exists surrounding the relationship between publications and reputation within the academic community. Higher rates of publication and publications with a higher impact factor are associated with more prestigious institutional reputations [2, 3].

Furthermore, publication output and the number of citations are primary factors when considering someone for promotions, job retention and mobility, and tenure [2], [3], [4, 29], [30], [31]. Within the world of academic medicine, the h-index is utilized as an objective measure to describe the impact that an individual’s work has made. The h-index is calculated utilizing both the number of publications and the number of citations by other authors per publication [29]. Higher rates of publication, and high h-index results, are associated with higher academic rank, NIH funding procurement, and career advancement for faculty physicians [2], [3], [4, 29], [30], [31]. Rates of high-impact publications are impossible to untangle from career advancement. Career advancement opportunities frequently offer additional financial incentives, further incentivizing prolific publication.

With numerous incentives to publish, academic physicians may attempt to fast-track the path to success by engaging in research misconduct or misrepresentation. DuBois et al. [32] investigated the motivation that is driving researchers to participate in research misconduct. The study found that 33 % felt pressure to publish, and 48 % of cases could be attributed to a self-centered personality such as having confidence they would not be caught, arrogance, seeking prestige, and greed [32]. More serious cases of fabrication, falsification, or plagiarism were seen more often with motivational factors of seniority, financial incentives, and oversight failures [32]. Although data for US physicians is lacking, a study conducted in 2000 in England showed that 59.8 % of physician survey respondents indicated that they felt pressured to publish to better their careers [1, 33]. It also revealed that 5.7 % of the physician respondents had participated in honorary authorship and 4.1 % would be willing to falsify data to improve a grant application [1, 33].

Consequences of research misconduct

There is not a universal process or consequence to research misconduct or misrepresentation in authorship. As a result, infractions such as research-related misconduct can result in a range of sanctions including article retraction, action by academic institutions or administrations, civil penalties, and criminal penalties [34]. Authorship disputes, including allegations of ghost authorship and others, are not viewed by the ORI as plagiarism, leaving journals and administrations to handle claims of misrepresentation as they see fit [34]. Journals have not yet enacted a universally accepted protocol for investigating and addressing research misrepresentation [11]. Many well-known journals, including Cell, Nature, and The Lancet, follow the guidelines proposed by the Committee on Publication Ethics (COPE) [35], [36], [37]. These guidelines encourage article retraction when necessary, along with publishing a notice of retraction outlining the reason [38]. A lack of government oversight and variability in journal response could be possible factors in the high and rising prevalence of authorship misrepresentation.

The ORI maintains lists of individuals that have committed research misconduct. Administrative actions imposed by the ORI are dependent upon the seriousness and impact of the misconduct, along with the individual’s history of behavior [39]. Possible sanctions are listed below:

… debarment from eligibility to receive Federal funds for grants and contracts; prohibition from service on PHS [Public Health Service] advisory committees, peer review committees, or as consultants; certification of information sources by respondent that is forwarded by institution; certification of data by institution; imposition of supervision on the respondent by the institution; submission of a correction of a published article by respondent; submission of a retraction of a published articles by respondent [39].

In the 1980s, a cardiology fellow was discovered to have fabricated data that led to false data being collected over a period of approximately 14 years. The author suffered a 10-year suspension from participating in federally funded research, along with the retraction of a total of 82 papers from two separate institutions [13]. This case is an example of a failure of the peer-review system and of oversight from the institution where the research was conducted. Reproducibility of data is incredibly important to avoid cases like this. In 2011, a study was conducted that reviewed articles retracted from PubMed between 2000 and 2010 [40]. Out of 742 articles, 73.5 % were retracted due to scientific error and 26.6 % due to fraud including falsification and fabrication of data [40]. A concerning finding of this study revealed that 31.8 % of the retracted articles did not have any indication that they had been retracted [40]. Both examples reveal the systemic gap that allows articles containing falsified and fraudulent material to remain available for continued citation by unsuspecting researchers. Utilizing information from retracted articles decreases the value of the work and the legitimacy of the author citing it.

Beyond the professional and legal consequences of research misconduct, ethical breaches in research can result in public harm and loss of confidence in the scientific community (Table 3). Perhaps the most socially impactful case of research misconduct is that of Andrew Wakefield, who published a paper in The Lancet in 1998 stating that the measles, mumps, and rubella (MMR) vaccine caused autism in children [41]. The author faced article retraction, banishment from practicing medicine, and worldwide attention from the media exposing him as a fraud. Despite proving that Wakefield falsified data and many studies affirming that there is no link between the MMR vaccine and autism, the social damage had already been done. Wakefield’s fabricated work scared the public, and MMR vaccination rates decreased. Decreased vaccination rates resulted in a resurgence of measles, which had previously been eliminated by the vaccine [41]. Not only did this led to the death of children in Britain and the United States, but also it started a movement of antivaccinators putting more children at risk for many preventable diseases [41]. The implications of research misconduct are far-reaching for individuals, institutions, the scientific community, and the greater public.

Potential solutions

Scientific research and publications serve as a conduit for furthering knowledge and understanding of biology and medicine. Publications are a tool to communicate research and motivate one’s peers [3]. Published works can initiate discussions regarding the interpretation of results, the application of the findings, and the generation of ideas for future works and advancement in knowledge. The development and execution of research studies require significant dedication and effort.

Within the realm of academic medicine, research misconduct and misrepresentation continue to occur without clear solutions. In the arena of clinical research, Bando et al. [4] propose a multifaceted approach to ensure research integrity. Although clinical research represents only one facet of academic medicine, their approach could be extrapolated to almost all areas of academic medical research and scholarly work. Their first recommendation involves the creation of a network of independent ORIs to work in tandem with local Institutional Review Boards (IRB). Although IRBs focus on ethics and the protection of human subjects, local ORIs would ensure the integrity of the scientific process [4]. Their second proposal involves creating and maintaining a local database to collect raw data related to clinical trials. This would allow for data review by a local ORI or journal editor, and data would be preserved and accessible to authorized parties for a minimum of 10 years [4]. The third element of the proposal calls for medical schools to implement the ORI’s responsible conduct of research (RCR) program into their curriculum [4]. This would ensure universal knowledge of ethical research practices by all US-trained physicians. The fourth and final proposal is for clinical researchers to implement a “Checklist for Clinical Research” [4]. Their proposed checklist implements checkpoints at each step of the project, from inception to publication, to ensure integrity throughout the process [4]. It also calls for transparency and collaboration among all members of the research team.

Although this comprehensive approach, if adopted, offers much-needed solutions to the problem of research misconduct in academic medicine, there are barriers to implementation. The most likely barrier for academic institutions to implement a comprehensive research integrity program is funding. The funds required to establish an independent ORI and a central database would be cost prohibitive for most nonprofit organizations, including universities, osteopathic medical schools, and hospitals. Other interventions, such as the implementation of a research integrity curriculum and a research integrity checklist system, are much more feasible. Other barriers that may be encountered are regulatory fatigue, department culture that resists buy-in, and time constraints for those tasked with implementation. Currently, osteopathic medical schools are required to include research methodology and research ethics in the curriculum but have no obligation to implement comprehensive research integrity systems [19]. Widespread and universal implementation of the changes proposed by Bando et al. into osteopathic education would likely require a mandate from Commission on Osteopathic College Accreditation (COCA), the accrediting body for osteopathic medical schools.

Limitations

This narrative review had limited access to full-length articles, which was a limitation. Additionally, only articles published after the year 2001 were included, limiting the historical context of this information. Narrative reviews are inherently limited by the completeness of a literature search. An inherent bias of the authors in their interpretation of content may also be present; however, this was attempted to be mitigated by including direct quotations from material and providing relevant information as complete as possible.