Abstract

Most biomedical journals now require authors to declare their conflicts of interest (COI), especially financial ones, before they accept the manuscript for submission. This study aims to examine the COI policies of Nepalese healthcare journals. The sample constituted journals indexed in Nepal Journals Online (NepJOL) as of June 2021. Of the 68 that met our inclusion criteria, 38(55.9%) journals endorsed the COI policy of the International Committee of Medical Journal Editors. Thirty-six (52.9%) journals had a policy for reporting the COI. Financial COI was the only type of COI mentioned. All journals in Nepal are encouraged to request the authors to declare the COI for better transparency.

Keywords: conflict of interest, healthcare journals, journal policy, Nepal, research ethics

Introduction

A conflict of interest (COI) exists when an individual or an institution has two or more overlapping, and often contradictory, interests in an activity. The Institute of Medicine, US, defines COI as “circumstances that create a risk that professional judgments or actions regarding a primary interest will be unduly influenced by a secondary interest” [1]. As the role of a clinician has widened with many serving as investors or part-time consultants in pharmaceutical companies and with the increasing involvement of pharmaceutical and other industries — such as tobacco, alcohol, food and beverage — in medical practice, the topic of COI is emerging as a powerful concept in research and publication ethics. Food and beverage industries often try to portray their products as healthy, whereas tobacco and alcohol industries try to undermine the ill effects of their products [2, 3, 4]. Financial COI is most common, though other forms of COI also exist [5, 6]. Financial COI includes but is not limited to stock ownership, grant/research support, direct employment, and serving as an advisor, consultant, or public advocate of the company [5]. Non-financial forms of COI include professional, intellectual, and personal conflicts [7]. Other types of COI, such as that of editors or reviewers, do exists but are not the subject of our study.

In healthcare research, COIs are widespread [8]. Studies have revealed that industry-funded research is more likely to derive conclusions in favour of the industry [9, 10, 11]. Given this context, the readers of a scholarly paper need to know the type and extent of the authors’ COI to interpret the results critically. Hence, declaring a COI appropriately and honestly is the responsibility of every author. Failure to do so not only erodes public trust in research reports but also misleads healthcare professionals, which may have dire consequences.

Journals have a unique responsibility of ensuring that the author makes appropriate disclosures. Many prestigious international organisations — such as the International Committee of Medical Journal Editors (ICMJE), the World Association of Medical Editors (WAME), and the Committee on Publication Ethics (COPE) — have published guidelines with specific recommendations stated explicitly on their website for the disclosure of information about authors’ COI [1, 12, 13, 14]. COPE is a subscription-based organisation that opines on case studies, whereas WAME is a discussion-based forum open to all editors. ICMJE, on the other hand, is an invitation-only committee that provides recommendations. Though there is a significant overlap, the three have different roles in promoting and enforcing publication ethics as well as different powers. COPE’s code of conduct recommends editors to have clear definitions of COI and processes for handling them [13]. WAME has specific guidelines regarding declaring and managing a COI [12]. Similarly, the ICMJE recommends publishing articles with statements declaring the authors’ relationships with the organisation, the source and nature of support, and the nature of involvement during the different phases of research. The ICMJE also encourages all journals endorsing its criteria to use the disclosure form developed by their committee [14]. Though none of the recommendations from these organisations is legally binding, they are considered important by the journal-editing community. Therefore, all journals are strongly advised to make their COI policies publicly available and readily accessible [12]. We believe that a journal’s COI guidelines should include a declaration, a mechanism of investigation, and potential action from the journal if an undisclosed COI is found. However, adherence to these ethical aspects varies across journals. Several studies evaluate the existence and content of the COI policies of journals across the world [6, 15, 16], but a similar study has not yet been published in Nepal.

Therefore, the primary objective of this study is to determine the proportion of healthcare journals published in Nepal that disclose their policies on their websites. The secondary objective is to describe their degree and pattern of adherence to standard COI policies as specified in the ICMJE guidelines [14].

Methods

Using the Nepal Journals Online (NepJOL) database (maintained by the Tribhuvan University Central Library in Nepal), we reviewed all healthcare-related journals published in the English language, as of first week of June 2021 [17]. NepJOL is a comprehensive database that lists all journals that fulfil a minimum set of standards and ranks them based on quality. NepJOL follows the Journal Publishing and Practices Standards (JPPS) criteria, designed specifically for developing countries. It assigns journals one of six levels (new title, inactive title, no stars, one star, two stars, and three stars) based on the publication of original research, a well-functioning editorial board, an accurate description of the peer-review process, the availability of authors’ guidelines, and the display of editorial and publishing policies. Details of the ranking criteria can be found on the JPPS website [https://www.journalquality.info/en/jpps-criteria/] [18].

As the information was publicly accessible, approval from the institutional review board (IRB) was not sought for this study. We identified 68 healthcare journals listed in NepJOL. All medical, dental, nursing public health colleges, and professional societies related to healthcare were enquired regarding the publication of an institutional journal. Altogether, 69 healthcare-related journals were found, of which, one was not listed in the NepJOL database, and hence, excluded from our study. We studied the “Information for Contributors” section of each journal in detail and analysed its COI guidelines. In particular, we looked at whether a journal mentioned a COI policy and what it consisted of. In journals that did not display this information or when the information was unclear, one of us registered as an author on their submission system to read the author’s guidelines to get the desired information.

The journals were divided based on their JPPS ranking, subject category, endorsement of COPE, WAME, or ICMJE, and journal affiliation to professional organisations or academic institutions. To retrieve data on COI disclosure requirements, we searched the page on information for the contributors for words or phrases such as “Conflict of Interest”, “financial disclosure”, “funded by”, “support”, “association”, “competing interests”, “relationship”, “employment status”, “affiliation”, and “acknowledgements”. The explanation of these words and phrases was analysed to find out the actual definition provided by each journal. Specifically, the following components were recorded:

• Presence or absence of a disclosure policy.

• If the journal had a COI disclosure policy, the type of COI mentioned (financial or others).

• If financial, the amount of payment and duration of employment or stock-holding.

• The journal’s policy regarding barring certain industry-related manuscripts (manuscripts discussing the benefits of food and beverages or the decreased harm of tobacco and alcohol).

• Policy description of how COI disclosure affects the manuscript review process.

• Policy description of how non-disclosure of COI affects the review process.

Though the last two points were not mentioned as standard components by WAME, COPE, and ICMJE, a previous report evaluating these among public health journals [5] prompted us to analyse these aspects in Nepalese journals as well.

For descriptive data presentation, we used frequencies and percentages to describe the categorical variables.

Results

A total of 226 journals were listed in the NepJOL database, of which 68 healthcare related journals met our inclusion criteria.

Journal ranking

Table 1 displays the ranking status of the journals on NepJOL. “Two stars” and “Working towards ranking” were the most common categories.

Table 1. Ranking of the journals in the NepJOL database

Rank	Number of journals (%)
Inactive	7 (10.3)
New	11 (16.2)
Working towards   ranking	17 (25)
One star	15 (22.1)
Two stars	18 (26.5)
Three stars	0
Total	68

Subject category of the journals

Table 2 shows the subject category of the journal. Multispecialty journals — covering all clinical fields like medicine, surgery, dentistry, nursing, basic sciences, etc — such as the Journal of Nepal Health Research Council (JNHRC), the Journal of Institute of Medicine Nepal (JIOM Nepal), and public health journals were the most common journals published in Nepal.

Table 2. Subject category of the journals in the NepJOL database

Subject category	Number of journals (%)
Multispecialty	35 (51.5)
Public Health	6 (8.8)
Dentistry	3 (4.4)
Laboratory Medicine	3 (4.4)
Neuroscience	3 (4.4)
Others/Subject-specific: One each from Anaesthesiology, Cardiology, Dermatology,   Endocrinology, General Surgery, Infectious Disease,   Internal Medicine, Obstetrics & Gynaecology, Oncology,   Ophthalmology, Orthopaedics, Otorhinolaryngology,   Pathology, Paediatrics, Pharmacy, Physiology, Psychiatry,   and Radiology	18 (26.5)
Total	68 (100)

Journal affiliation

Thirty journals were published by professional societies, 19 by individual medical colleges, and seven by universities and academies. The remaining were published by departments of universities and colleges, individual hospitals, or nongovernmental organisations.

Endorsement of COPE, WAME, or ICMJE

Thirty-eight (55.9%) journals endorsed the ICMJE policies, 20 (29.4%) endorsed the COPE, and 13 (19.1%) endorsed policies that of WAME. Only 12 (17.6%) journals endorsed the policy of all three organisations. One journal endorsed the policy of WAME only, six journals endorsed the policy of both ICMJE and WAME. Twenty-nine journals did not endorse any of these organisations’ policies.

COI statement policy

Out of the 68 journals, 36 (52.9%) had a COI policy explicitly stated on the “Information for Contributors” page. Table 3 shows the number and percentage of journals with a COI policy for each rating.

Table 3. Journals providing a COI policy and their ratings (n = 36)

Ratings	Number (%)
New and inactive	8/18 (44.4)
Working towards   stars	10/17 (58.8%)
One star	7/15 (46.7%)
Two stars	11/18 (61.1%)
Total	36/68 (52.9%)

Of the 38 journals that endorsed ICMJE guidelines, 13 (34.2%) did not have a COI statement. Whereas, of the 30 journals that did not endorse ICMJE guidelines, 13 (43.3%) had a COI statement.

Table 4 shows the relationship between the subject category of journals and the COI policy statement. Forty-four journals that have published at least two journals in one category were analysed.

Table 4. Subject category of the journals and the COI policy statement (n = 44)

Journal Type	Number (%)
Multispecialty	20/ 35 (57.1)
Public Health	2/6 (33.3)
Dentistry	1/3 (33.3)
Total	23/44 (52.3)

Only 21/36 (55.3%) journals further elaborated on COI with examples. All 36 journals mentioned only financial COI in their COI description. Only four (11.1%) journals mentioned that the manuscript would not be sent to a peer reviewer working in the same organisation as the author.

Out of 68, only one (1.5%) journal required the authors to mention the amount and duration of funding. ICMJE recommends authors to declare the support received for the ongoing manuscript without any time limit, whereas the time frame for the disclosure of items such as grants, royalties, consulting fees, and honoraria is 36 months [14].

None of the journals mentioned the degree of independence from the funders, while drafting and submitting the manuscript. No journal mentioned any policy regarding manuscripts funded by certain industries such as tobacco, alcohol, food and beverage. None of the journals had any description of non-disclosure of COI affecting the review process.

Discussion

Relationships between researchers and manufacturing companies such as those of drugs, devices, and tests have the potential to be exploited such that the financial and other interests of the researcher may conflict with the pure aim of advancing medicine [20]. To strike a balance, IRBs and data safety and monitoring boards (DSMBs) regulate the COI during the research process while journals are expected to evaluate the COI during the publication process [21, 22]. Hence, many leading medical journals adopt a policy of authors declaring the degree and magnitude of COI [5, 6, 15]. Multiple studies published in Europe and North America have explored journals’ adherence to COI policies [6, 7, 8, 16]. Data on this topic specifically from South Asian journals are scarce [20, 23]. This is the first study in Nepal that investigates this important part of the publication process in 68 healthcare journals across more than 21 healthcare specialties.

General characteristics of the journals

In our study, 12 (17.6%) of the 68 journals mentioned that they followed the guidelines of all three organisations — COPE, WAME, and ICMJE. Thirty-eight (55.9%) journals mentioned that they followed the ICMJE guidelines, 13 (19.1%) endorsed WAME guidelines, and 20 (29.4%) endorsed those of COPE. In the study of public health journals by Daou et al [5], only 21% of the journals were members of the ICMJE and 67% were members of COPE. While in a study of 256 high-impact journals by Blum et al [6], only 69 (26.9%) journals had officially endorsed the ICMJE guidelines. Most journals endorsing ICMJE guidelines had COI policies compared to the 84% of journals not endorsing the guidelines (68/69 versus 158/187) [6].

Dal-Ré et al in 2018 studied “highly influential” journals based on the 2017 Journal Citation Reports [24]. They found that 45% (58/130) of these journals followed the ICMJE recommendations and 73% (95/130) were COPE members. There was no correlation between the membership of these organisations and the display of the COI policy. Bose et al, in their analysis of 106 Indian biomedical journals, found that 61 (57.5%) journals endorsed all three international guidelines [23].

COI statement policy

In our study, 52.9% of the journals displayed COI policies on their website, which is smaller in comparison to Blum et al where 89% of medical journals had COI policies [6]. Daou et al found that 90% journals had a policy of reporting funding information [5]. This is in contrast to results obtained by Ancker et al, where only 33% (28/84) journals reported COI [16]. However, this study included journals from nonmedical subjects as well. Similarly, in another research of 53 bioethics journals by Master et al, 57% had COI policies for authors [25].

The star rating by JPPS criteria is given only when a journal meets a certain standard and has strong ethical policies [18]. Moreover, the ranking system of JPPS does not seem to incorporate COI policies while categorising the journals. Only three-star journals require the disclosure of funding acknowledgement [18].

Elaboration of the COI policy

In our study, of 36 journals, 21 described the actual meaning of COI, though this is surprisingly low compared to the published literature. In Blum et al, 77% had defined the meaning of COI [6]. While only 33% of journals adequately explained COI in biomedical publications from India as per Bose et al [23].

Discussions on the amount and duration of financial support are even more scarce. This is especially important if the study is related to the advancement of a new device or product. Daou et al found that none of the public health journals studied by them described the amount and duration of financial support [5]. In our study, only one journal described the amount and duration.

Effect of disclosure/non-disclosure of the COI on the manuscript evaluation process

Biomedical journals have largely remained silent regarding the effect of a disclosure of a COI on the editorial process. No journal in our study shared this information. In 2017, Daou et al found that 19% of the public health journals described the effect of COI disclosures on the editorial process [5]. The authors also noted that 10% of the journals mentioned a potential impact of not disclosing COI on the editorial process (eg, editors may correct, reject, or retract the publication). According to Cooper et al, 11% of the journals had a policy of restricting author submissions based on COI [15]. Master et al found that 31% of the journals specified that depending on the nature of the COI, the article might not be published [25]. They also found 17% of journals with a COI policy that stated certain penalties for authors for non-compliance, such as retraction of the article.

Regarding autonomy of publication, everyone believes that the authors should be as independent as possible, although the pressure from big industries can be huge [9, 10, 26]. For example, some industries require that the final manuscript be approved by them before submitting it to the journal. In our study, we found that none of the journals required the authors to disclose the contract between the authors and funding agencies.

Recommendations for the journal editors

Our study provides new data on the availability and content of the COI policies of the healthcare journals in Nepal. Though the findings are not discouraging, there is room for improvement. We recommend the following based on our findings:

1. Journals should explicitly state their COI policy on their website. The policy should clearly state how will the COI be verified and what will be the fate of the paper if the authors are found to state the COI dishonestly.

2. There should be some uniformity across all journals in terms of the minimum information required by the authors to indicate COI. Formulation of guidelines by the national society of healthcare journal editors will be a welcome step in this regard.

Limitations

We analysed journals listed only on NepJOL. Among those, only one healthcare journal was not included in the analysis.

Conclusion

This study reflects the COI policy of nearly 99% of healthcare-related journals from Nepal. Slightly more than half have a policy regarding authors’ COI declaration. However, many do not provide details on COI and the mechanisms to tackle nondisclosure or dishonest disclosure of COI in the “Information for Contributors” section. More healthcare journals from Nepal should have a policy requiring transparent reporting of COIs so that journals comply with international standards. This article concentrated only on the authors’ COI. In future, comprehensive studies on COI involving authors, editors, and reviewers should be conducted. Also, though inclusion is not mandatory, all Nepalese journals are encouraged to apply for enlistment on the NepJOL database to increase their visibility.

Conflict of Interest:

1) Mohan Raj Sharma: Ex-editor-in-chief of Nepal Journal of Neuroscience; Ex-executive editor of Journal of Institute of Medicine, Nepal; Editor of Journal of Neurosurgery in Rural Practice; Chairperson of Institutional Review Committee, Institute of Medicine, Nepal.

2) Lava Shrestha: Managing editor of the Journal of Institute of Medicine Nepal; Editor of Journal of Physiology Society of Nepal and Journal of Nepal Health Research Council.

3) Gopal Sedain: Ex-editor of Nepal Journal of Neurosciences.

4) Subigya Parajuli and Prakash Kafle: None declared.

Source of funding: None declared.